mirror of
https://github.com/waybarrios/opencode-power-pack.git
synced 2026-11-19 02:16:19 +00:00
Update the website for OpenCode 2 and all 55 skills
This commit is contained in:
+16
-6
@@ -8,10 +8,16 @@ Use the selective npm installer for Codex, OpenCode, or Pi when the user wants o
|
||||
|
||||
Use Claude Code's native plugin commands for the complete namespaced catalog. Do not assume that Claude Code loads the portable `.agents/skills` path.
|
||||
|
||||
## Current source and OpenCode 2.0
|
||||
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
The project-local procedure below uses the published npm catalog. Read its output before choosing skills. Do not try to install humanizer from npm 0.5.0 or change sources without the user's selection.
|
||||
|
||||
## Safe project-local procedure
|
||||
|
||||
1. Confirm the current directory belongs to the intended Git repository.
|
||||
2. Run `node --version`. Stop if Node.js is older than 20.
|
||||
2. Run `node --version`. Stop if Node.js is older than 20.11.0.11.0.
|
||||
3. Preview the resolved selection and destination.
|
||||
4. Report the destination, selected skills, automatic companions, existing conflicts, and planned writes.
|
||||
5. If the preview matches the user's requested scope, repeat the command without `--dry-run`.
|
||||
@@ -56,8 +62,9 @@ Verify through `/plugin`, then invoke a workflow with its collision-safe namespa
|
||||
|
||||
## Host verification
|
||||
|
||||
- Codex: start a new session, inspect `/plugins`, and invoke `$code-review` or another installed ID.
|
||||
- OpenCode: start a new session, run `opencode debug skill`, and invoke `/code-review` or another installed ID.
|
||||
- Codex: start a new session and invoke `$code-review` or another installed ID. Inspect `/plugins` only for a native plugin installation.
|
||||
- OpenCode 1: start a new session, run `opencode debug skill`, and invoke `/code-review` or another installed ID.
|
||||
- OpenCode 2: start a new session, check the skill picker, and invoke `/code-review` or another installed ID. Use `opencode debug agents` only to inspect specialists from the native plugin.
|
||||
- Pi: start a new session, run `pi list`, and ask for a task matching the installed workflow.
|
||||
- Claude Code: inspect `/plugin` and use `/opencode-power-pack:<skill-id>`.
|
||||
|
||||
@@ -68,6 +75,8 @@ Help me choose and install the right OpenCode Power Pack skills safely and proje
|
||||
|
||||
Your goal is to understand what I need, recommend the smallest useful profile or set of individual skills, obtain my explicit selection, install it for the active coding-agent host, preserve all existing project work, and leave a reproducible verification report.
|
||||
|
||||
The npm procedure below uses the published catalog. OpenCode 2 can use those copied skills, but the native V2 plugin and humanizer require Git main. If I need either, show the current Git install guide and wait for my source choice before continuing.
|
||||
|
||||
Safety rules:
|
||||
1. Do not modify package.json, package-lock.json, npm-shrinkwrap.json, pnpm-lock.yaml, yarn.lock, bun.lock, or application dependencies.
|
||||
2. Do not use --force, --global, --all, sudo, destructive Git commands, or manual deletion/replacement of existing skill directories. Do not choose a profile or skill for me, and do not install anything before I explicitly approve the selection.
|
||||
@@ -88,7 +97,7 @@ Phase 2: preflight without changing repository files.
|
||||
|
||||
1. Print the current directory with pwd.
|
||||
2. Run git rev-parse --show-toplevel. Stop if this is not a Git repository. Save the absolute Git root.
|
||||
3. Run node --version. Parse the major version and require Node.js 20 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
3. Run node --version. Parse the full version and require Node.js 20.11.0 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
4. Run npm --version and stop if npm or npx is unavailable.
|
||||
5. Capture git status --short before installation. Use it only as a baseline; do not alter any listed file.
|
||||
6. State that the only permitted destination is <git-root>/.agents/skills.
|
||||
@@ -130,8 +139,9 @@ Phase 6: verify and report.
|
||||
3. Run git status --short again and compare it with the saved baseline. Confirm that installation introduced no changes outside .agents/skills and did not change dependency manifests or lockfiles. Do not revert unrelated differences.
|
||||
4. Tell the user to start a new agent session if the skills do not appear automatically.
|
||||
5. Give the host-specific verification command:
|
||||
- Codex: inspect /plugins, then invoke an installed skill such as $code-review.
|
||||
- OpenCode: run opencode debug skill, then invoke /code-review.
|
||||
- Codex: invoke an installed skill such as $code-review in a new session. Use /plugins only for a native plugin installation.
|
||||
- OpenCode 1: run opencode debug skill, then invoke /code-review.
|
||||
- OpenCode 2: check the skill picker, then invoke /code-review.
|
||||
- Pi: run pi list, then request a task that matches an installed workflow.
|
||||
6. Finish with a concise structured report containing: stated need, approved selection and rationale, alternatives not chosen, host, Node/npm versions, Git root, destination, exact commands run, resolved skill count and names, installed entries, preserved/skipped entries, warnings or failures, worktree comparison, verification result, and any user action still required.
|
||||
```
|
||||
|
||||
Binary file not shown.
|
Before Width: | Height: | Size: 67 KiB After Width: | Height: | Size: 39 KiB |
@@ -12,11 +12,11 @@
|
||||
<path d="M145 0 0 190h103L78 335l170-219H138z" fill="#ffd60a"/>
|
||||
</g>
|
||||
<g font-family="IBM Plex Mono,monospace">
|
||||
<text x="390" y="150" fill="#9aa6b2" font-size="22" font-weight="700" letter-spacing="5">OPEN SOURCE // v0.5.0</text>
|
||||
<text x="390" y="150" fill="#9aa6b2" font-size="22" font-weight="700" letter-spacing="5">OPEN SOURCE // MAIN</text>
|
||||
<text x="390" y="270" fill="#f1f0eb" font-family="Barlow Condensed,sans-serif" font-size="94" font-weight="800" letter-spacing="-2">OPENCODE</text>
|
||||
<text x="390" y="370" fill="#ffd60a" font-family="Barlow Condensed,sans-serif" font-size="86" font-weight="800" letter-spacing="2">POWER PACK</text>
|
||||
<text x="390" y="440" fill="#f1f0eb" font-size="28" font-weight="600">54 SKILLS · FOUR CODING AGENTS</text>
|
||||
<text x="390" y="492" fill="#9aa6b2" font-size="20">CLAUDE CODE + CODEX + OPENCODE + PI</text>
|
||||
<text x="390" y="440" fill="#f1f0eb" font-size="28" font-weight="600">55 SKILLS · FOUR CODING AGENTS</text>
|
||||
<text x="390" y="492" fill="#9aa6b2" font-size="20">CLAUDE CODE + CODEX + OPENCODE 1 + 2 + PI</text>
|
||||
</g>
|
||||
<g transform="translate(390 540)" font-family="IBM Plex Mono,monospace" font-size="16" font-weight="700">
|
||||
<rect width="500" height="42" fill="#111821" stroke="#263340"/>
|
||||
|
||||
|
Before Width: | Height: | Size: 1.7 KiB After Width: | Height: | Size: 1.7 KiB |
+108
-66
@@ -21,7 +21,7 @@
|
||||
<meta property="og:title" content="OpenCode Power Pack Documentation">
|
||||
<meta
|
||||
property="og:description"
|
||||
content="Installation and usage guides for 54 Agent Skills across Claude Code, Codex, OpenCode, and Pi."
|
||||
content="Installation and usage guides for 55 Agent Skills across Claude Code, Codex, OpenCode, and Pi."
|
||||
>
|
||||
<meta property="og:url" content="https://skills.waybarrios.com/docs/">
|
||||
<meta
|
||||
@@ -34,7 +34,7 @@
|
||||
<meta property="og:image:alt" content="OpenCode Power Pack installation and operations documentation">
|
||||
<meta name="twitter:card" content="summary_large_image">
|
||||
<meta name="twitter:title" content="OpenCode Power Pack Documentation">
|
||||
<meta name="twitter:description" content="Install and use 54 Agent Skills across Claude Code, Codex, OpenCode, and Pi.">
|
||||
<meta name="twitter:description" content="Install and use 55 Agent Skills across Claude Code, Codex, OpenCode, and Pi.">
|
||||
<meta name="twitter:image" content="https://skills.waybarrios.com/assets/social-card.png">
|
||||
<meta name="twitter:image:alt" content="OpenCode Power Pack installation and operations documentation">
|
||||
<meta name="twitter:image:width" content="1200">
|
||||
@@ -45,51 +45,58 @@
|
||||
href="https://fonts.googleapis.com/css2?family=Barlow+Condensed:wght@600;700;800&family=IBM+Plex+Mono:wght@400;500;600;700&display=swap"
|
||||
rel="stylesheet"
|
||||
>
|
||||
<link rel="stylesheet" href="../styles.css?v=0.5.0-sandbox-1">
|
||||
<link rel="stylesheet" href="../styles.css?v=2026-10-19-opencode-2">
|
||||
<script type="application/ld+json">
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@type": "TechArticle",
|
||||
"headline": "OpenCode Power Pack Documentation",
|
||||
"description": "Installation, usage, and native sandbox documentation for OpenCode Power Pack Agent Skills.",
|
||||
"url": "https://skills.waybarrios.com/docs/",
|
||||
"dateModified": "2026-10-19",
|
||||
"author": {
|
||||
"@type": "Person",
|
||||
"name": "Wayner Barrios",
|
||||
"url": "https://github.com/waybarrios"
|
||||
},
|
||||
"about": {
|
||||
"@type": "SoftwareSourceCode",
|
||||
"name": "OpenCode Power Pack",
|
||||
"codeRepository": "https://github.com/waybarrios/opencode-power-pack",
|
||||
"softwareVersion": "main",
|
||||
"runtimePlatform": [
|
||||
"Claude Code",
|
||||
"Codex",
|
||||
"OpenCode 1 and 2",
|
||||
"Pi"
|
||||
]
|
||||
},
|
||||
"breadcrumb": {
|
||||
"@type": "BreadcrumbList",
|
||||
"itemListElement": [
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@type": "TechArticle",
|
||||
"headline": "OpenCode Power Pack Documentation",
|
||||
"description": "Installation, usage, and native sandbox documentation for OpenCode Power Pack Agent Skills.",
|
||||
"url": "https://skills.waybarrios.com/docs/",
|
||||
"dateModified": "2026-08-17",
|
||||
"author": {
|
||||
"@type": "Person",
|
||||
"name": "Wayner Barrios",
|
||||
"url": "https://github.com/waybarrios"
|
||||
},
|
||||
"about": {
|
||||
"@type": "SoftwareSourceCode",
|
||||
"name": "OpenCode Power Pack",
|
||||
"codeRepository": "https://github.com/waybarrios/opencode-power-pack"
|
||||
},
|
||||
"breadcrumb": {
|
||||
"@type": "BreadcrumbList",
|
||||
"itemListElement": [
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"name": "OpenCode Power Pack",
|
||||
"item": "https://skills.waybarrios.com/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"name": "Documentation",
|
||||
"item": "https://skills.waybarrios.com/docs/"
|
||||
}
|
||||
]
|
||||
}
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"name": "OpenCode Power Pack",
|
||||
"item": "https://skills.waybarrios.com/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"name": "Documentation",
|
||||
"item": "https://skills.waybarrios.com/docs/"
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
</script>
|
||||
</head>
|
||||
<body class="docs-body">
|
||||
<a class="skip-link" href="#docs-main">Skip to documentation</a>
|
||||
<div class="status-rail" aria-hidden="true">
|
||||
<span>DOCUMENTATION // v0.5.0</span>
|
||||
<span>54 MODULES</span>
|
||||
<span>LAST UPDATED 2026-08-17</span>
|
||||
<span>DOCUMENTATION // MAIN</span>
|
||||
<span>55 MODULES</span>
|
||||
<span>LAST UPDATED 2026-10-19</span>
|
||||
</div>
|
||||
|
||||
<div class="docs-shell">
|
||||
@@ -101,6 +108,7 @@
|
||||
<span><b>POWER PACK</b> DOCS</span>
|
||||
</a>
|
||||
<nav aria-label="Documentation sections">
|
||||
<a href="#opencode-2">OpenCode 2.0 + updates</a>
|
||||
<a href="#quick-start">Choose an install</a>
|
||||
<a href="#npm">Selective npm</a>
|
||||
<a href="#agent-install">Agent-assisted install</a>
|
||||
@@ -109,7 +117,7 @@
|
||||
<a href="#how-skills-work">How skills work</a>
|
||||
<a href="#hosts">Full host installs</a>
|
||||
<a href="#profiles">Profiles + contents</a>
|
||||
<a href="skills/">All 54 skills</a>
|
||||
<a href="skills/">All 55 skills</a>
|
||||
<a href="#reviews">Review skills</a>
|
||||
<a href="#maintenance">Update and remove</a>
|
||||
<a href="#troubleshooting">Troubleshooting</a>
|
||||
@@ -133,18 +141,26 @@
|
||||
</p>
|
||||
</header>
|
||||
|
||||
<section class="docs-section" id="opencode-2">
|
||||
<h2>OpenCode 2.0 and current source</h2>
|
||||
<p>The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.</p>
|
||||
<p>The npm installer can copy its released skills for OpenCode 2. Copying skills does not register the three named specialists.</p>
|
||||
<p>Smoke tests cover OpenCode 1.18.7, 2.0.14, and 2.0.32. The current suite has 418 tests; the latest local run passed 417 and skipped the optional native sandbox integration.</p>
|
||||
<p>For the host commands and permission checks, see <a href="#host-opencode">OpenCode setup</a>. For the new writing workflow, see <a href="skills/#humanizer">humanizer</a>.</p>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="quick-start">
|
||||
<h2>Choose an installation path</h2>
|
||||
<p>The two paths solve different problems. Choose one before running commands.</p>
|
||||
<p>Choose selected skills from npm or the full plugin from Git main. The sources contain different versions.</p>
|
||||
<div class="install-choice-grid">
|
||||
<article>
|
||||
<span>SELECTIVE</span>
|
||||
<h3>npm installer</h3>
|
||||
<p>Best when Codex, OpenCode, or Pi should load only the workflows you select.</p>
|
||||
<ul>
|
||||
<li>Install one skill, several skills, a profile, or all 54.</li>
|
||||
<li>Install one skill, several skills, a profile, or the released catalog.</li>
|
||||
<li>Choose user-level or repository-local scope.</li>
|
||||
<li>Requires Node.js 20 or newer.</li>
|
||||
<li>Requires Node.js 20.11.0 or newer.</li>
|
||||
</ul>
|
||||
<div class="docs-code"><code>npx @waybarrios/opencode-power-pack install --profile recommended</code></div>
|
||||
<a href="#npm">Follow the selective guide →</a>
|
||||
@@ -152,7 +168,7 @@
|
||||
<article>
|
||||
<span>COMPLETE CATALOG</span>
|
||||
<h3>Native host plugin</h3>
|
||||
<p>Best when you want all 54 workflows through the host's normal package manager.</p>
|
||||
<p>Best when you want all 55 workflows through the host's normal package manager.</p>
|
||||
<ul>
|
||||
<li>Claude Code receives collision-safe namespaced skills.</li>
|
||||
<li>Codex, OpenCode, and Pi receive native package integration.</li>
|
||||
@@ -177,7 +193,7 @@ pi install ...</code></div>
|
||||
<ol class="installation-steps">
|
||||
<li>
|
||||
<h3>Inspect the catalog</h3>
|
||||
<p>This prints the eight profiles and all 54 available skill IDs. It does not modify files.</p>
|
||||
<p>This prints the profiles and skill IDs in the published npm package. It does not modify files.</p>
|
||||
<div class="docs-code"><code>npx @waybarrios/opencode-power-pack list</code></div>
|
||||
</li>
|
||||
<li>
|
||||
@@ -194,7 +210,7 @@ npx @waybarrios/opencode-power-pack install --profile review
|
||||
|
||||
# Combine a profile with another skill
|
||||
npx @waybarrios/opencode-power-pack install --profile frontend code-quality</code></div>
|
||||
<p><a href="skills/">Compare all 54 skills and copy individual commands →</a></p>
|
||||
<p><a href="skills/">Compare all 55 skills and copy individual commands →</a></p>
|
||||
</li>
|
||||
<li>
|
||||
<h3>Choose the destination</h3>
|
||||
@@ -231,6 +247,15 @@ npx @waybarrios/opencode-power-pack install --profile security</code></div>
|
||||
</p>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="github-packages">
|
||||
<h2>GitHub Packages mirror</h2>
|
||||
<p>npmjs is the public install source and does not require GitHub authentication. The released package is also available through GitHub Packages.</p>
|
||||
<p>For the mirror, use a GitHub personal access token (classic) with <code>read:packages</code>:</p>
|
||||
<div class="docs-code"><code>npm login --scope=@waybarrios --auth-type=legacy --registry=https://npm.pkg.github.com
|
||||
npm install --global @waybarrios/opencode-power-pack --registry=https://npm.pkg.github.com</code></div>
|
||||
<p>The mirror follows release tags. It does not include additions that are only on main.</p>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="agent-install">
|
||||
<h2>Let your agent install it</h2>
|
||||
<p>
|
||||
@@ -246,6 +271,8 @@ npx @waybarrios/opencode-power-pack install --profile security</code></div>
|
||||
|
||||
Your goal is to understand what I need, recommend the smallest useful profile or set of individual skills, obtain my explicit selection, install it for the active coding-agent host, preserve all existing project work, and leave a reproducible verification report.
|
||||
|
||||
The npm procedure below uses the published catalog. OpenCode 2 can use those copied skills, but the native V2 plugin and humanizer require Git main. If I need either, show the current Git install guide and wait for my source choice before continuing.
|
||||
|
||||
Safety rules:
|
||||
1. Do not modify package.json, package-lock.json, npm-shrinkwrap.json, pnpm-lock.yaml, yarn.lock, bun.lock, or application dependencies.
|
||||
2. Do not use --force, --global, --all, sudo, destructive Git commands, or manual deletion/replacement of existing skill directories. Do not choose a profile or skill for me, and do not install anything before I explicitly approve the selection.
|
||||
@@ -266,7 +293,7 @@ Phase 2: preflight without changing repository files.
|
||||
|
||||
1. Print the current directory with pwd.
|
||||
2. Run git rev-parse --show-toplevel. Stop if this is not a Git repository. Save the absolute Git root.
|
||||
3. Run node --version. Parse the major version and require Node.js 20 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
3. Run node --version. Parse the full version and require Node.js 20.11.0 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
4. Run npm --version and stop if npm or npx is unavailable.
|
||||
5. Capture git status --short before installation. Use it only as a baseline; do not alter any listed file.
|
||||
6. State that the only permitted destination is <git-root>/.agents/skills.
|
||||
@@ -308,8 +335,9 @@ Phase 6: verify and report.
|
||||
3. Run git status --short again and compare it with the saved baseline. Confirm that installation introduced no changes outside .agents/skills and did not change dependency manifests or lockfiles. Do not revert unrelated differences.
|
||||
4. Tell the user to start a new agent session if the skills do not appear automatically.
|
||||
5. Give the host-specific verification command:
|
||||
- Codex: inspect /plugins, then invoke an installed skill such as $code-review.
|
||||
- OpenCode: run opencode debug skill, then invoke /code-review.
|
||||
- Codex: invoke an installed skill such as $code-review in a new session. Use /plugins only for a native plugin installation.
|
||||
- OpenCode 1: run opencode debug skill, then invoke /code-review.
|
||||
- OpenCode 2: check the skill picker, then invoke /code-review.
|
||||
- Pi: run pi list, then request a task that matches an installed workflow.
|
||||
6. Finish with a concise structured report containing: stated need, approved selection and rationale, alternatives not chosen, host, Node/npm versions, Git root, destination, exact commands run, resolved skill count and names, installed entries, preserved/skipped entries, warnings or failures, worktree comparison, verification result, and any user action still required.</code></div>
|
||||
<button
|
||||
@@ -352,7 +380,7 @@ npx @waybarrios/opencode-power-pack install --profile recommended --project</cod
|
||||
<tr><td><code>list</code></td><td>Print all profiles and skill IDs from the installed package.</td></tr>
|
||||
<tr><td><code>install <skill...></code></td><td>Install one or more explicit skill IDs.</td></tr>
|
||||
<tr><td><code>--profile <name></code></td><td>Add a curated profile. Repeat the option to combine profiles.</td></tr>
|
||||
<tr><td><code>--all</code></td><td>Select the complete 54-skill catalog.</td></tr>
|
||||
<tr><td><code>--all</code></td><td>Select every skill in the installed package.</td></tr>
|
||||
<tr><td><code>--global</code></td><td>Write to <code>~/.agents/skills</code>. This is the default.</td></tr>
|
||||
<tr><td><code>--project</code></td><td>Find the current Git root and write to <code>.agents/skills</code> inside it.</td></tr>
|
||||
<tr><td><code>--dry-run</code></td><td>Show the resolved destination and changes without writing files.</td></tr>
|
||||
@@ -548,7 +576,7 @@ Do not run the child command separately and do not retry without the sandbox.</c
|
||||
<section class="docs-section" id="hosts">
|
||||
<h2>Full native host installs</h2>
|
||||
<p>
|
||||
Each method below installs all 54 workflows. Use the npm path above when you want a smaller
|
||||
Each method below installs all 55 workflows. Use the npm path above when you want a smaller
|
||||
set for Codex, OpenCode, or Pi.
|
||||
</p>
|
||||
|
||||
@@ -578,14 +606,20 @@ codex plugin list --marketplace opencode-power-pack</code></div>
|
||||
</article>
|
||||
|
||||
<article class="host-doc" id="host-opencode">
|
||||
<header><span>03</span><h3>OpenCode</h3><b>FULL CATALOG</b></header>
|
||||
<p>OpenCode 1.18.7 or newer is required. Pin v0.5.0 for a reproducible installation.</p>
|
||||
<div class="docs-code"><code>opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git#v0.5.0"</code></div>
|
||||
<header><span>03</span><h3>OpenCode 1 and 2</h3><b>FULL CATALOG</b></header>
|
||||
<p>Use Git main for OpenCode 2.0 and all 55 skills. The v0.5.0 tag predates the V2 plugin port.</p>
|
||||
<div class="docs-code"><code># OpenCode 2
|
||||
opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
|
||||
# OpenCode 1
|
||||
opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"</code></div>
|
||||
<dl>
|
||||
<div><dt>Restart</dt><dd>Start a new OpenCode session after the command finishes.</dd></div>
|
||||
<div><dt>Verify</dt><dd>Run <code>opencode debug skill</code>; it should list all 54 unprefixed IDs.</dd></div>
|
||||
<div><dt>Invoke</dt><dd>Use the generated slash command, for example <code>/code-review</code>.</dd></div>
|
||||
<div><dt>Restart</dt><dd>Start a new OpenCode session after installation.</dd></div>
|
||||
<div><dt>Verify V2</dt><dd>Run <code>opencode debug agents</code> and check the skill picker for all 55 workflows.</dd></div>
|
||||
<div><dt>Verify V1</dt><dd>Run <code>opencode debug skill</code> and <code>opencode debug agent code-explorer</code>.</dd></div>
|
||||
<div><dt>Invoke</dt><dd>Use a slash command such as <code>/code-review</code>, <code>/feature-dev</code>, or <code>/humanizer</code>.</dd></div>
|
||||
</dl>
|
||||
<p>V2 evaluates the bundled specialists' read-only policy after global rules. The agent listing can show the host's base permissions. A custom system prompt or primary mode replaces the bundled role and keeps the host's permission defaults.</p>
|
||||
</article>
|
||||
|
||||
<article class="host-doc" id="host-pi">
|
||||
@@ -661,11 +695,11 @@ pi install git:github.com/waybarrios/opencode-power-pack -l</code></div>
|
||||
</div>
|
||||
</details>
|
||||
<details>
|
||||
<summary><span><code>authoring</code> Reusable workflow creation</span><b>3 skills</b></summary>
|
||||
<summary><span><code>authoring</code> Reusable workflow creation</span><b>4 on main</b></summary>
|
||||
<div>
|
||||
<p>Create Agent Skills and MCP servers, or turn technical papers into engineering-ready summaries.</p>
|
||||
<div class="docs-code"><code>npx @waybarrios/opencode-power-pack install --profile authoring</code></div>
|
||||
<p><code>mcp-builder</code>, <code>paper-summarizer</code>, and <code>skill-creator</code>.</p>
|
||||
<div class="docs-code"><code>npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install --profile authoring</code></div>
|
||||
<p>On main: <code>humanizer</code>, <code>mcp-builder</code>, <code>paper-summarizer</code>, and <code>skill-creator</code>. npm 0.5.0 has the three earlier skills. <a href="skills/#humanizer">Install humanizer from source.</a></p>
|
||||
</div>
|
||||
</details>
|
||||
<details>
|
||||
@@ -677,7 +711,7 @@ pi install git:github.com/waybarrios/opencode-power-pack -l</code></div>
|
||||
</div>
|
||||
</details>
|
||||
</div>
|
||||
<p><a href="skills/">Open the complete 54-skill reference for individual installation commands →</a></p>
|
||||
<p><a href="skills/">Open the complete 55-skill reference for individual installation commands →</a></p>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="reviews">
|
||||
@@ -736,19 +770,27 @@ npm uninstall --global @waybarrios/opencode-power-pack</code></div>
|
||||
codex plugin add opencode-power-pack@opencode-power-pack
|
||||
codex plugin remove opencode-power-pack@opencode-power-pack</code></div>
|
||||
|
||||
<h3>OpenCode</h3>
|
||||
<div class="docs-code"><code>opencode plugin --global --force "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"</code></div>
|
||||
<p>To remove it, delete the matching <code>opencode-power-pack@...</code> entry from the config where it was installed, then restart OpenCode.</p>
|
||||
<h3>OpenCode 2</h3>
|
||||
<div class="docs-code"><code>opencode plugin update "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
opencode plugin remove "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"</code></div>
|
||||
<p>Use the exact configured specification, including a tag or local Git URL. On OpenCode 1, update by reinstalling with <code>plugin --global</code>; remove its entry from the <code>plugin</code> config array to uninstall. For local clones, pull the source and then update or reinstall.</p>
|
||||
|
||||
<h3>Pi</h3>
|
||||
<div class="docs-code"><code>pi update git:github.com/waybarrios/opencode-power-pack
|
||||
pi remove git:github.com/waybarrios/opencode-power-pack</code></div>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="evaluations">
|
||||
<h2>Contributor checks and live evaluations</h2>
|
||||
<p>The behavioral runner uses the OpenCode 2 session API. It rejects partial text from failed or interrupted turns and accepts optional <code>provider/model#variant</code> model references.</p>
|
||||
<p>Live runs need provider environment credentials. The isolated runtime does not copy OpenCode 2's credential database. See the <a href="https://github.com/waybarrios/opencode-power-pack/blob/main/evals/behavioral/README.md">evaluation guide</a> for setup and evidence review.</p>
|
||||
<p>The <a href="https://github.com/waybarrios/opencode-power-pack/blob/main/docs/opencode-compatibility.md">OpenCode compatibility guide</a> covers version checks, source selection, updates, and specialist permissions.</p>
|
||||
</section>
|
||||
|
||||
<section class="docs-section" id="troubleshooting">
|
||||
<h2>Troubleshooting</h2>
|
||||
<h3>Node rejects the package</h3>
|
||||
<p>The selective installer requires Node.js 20 or newer. Check with <code>node --version</code>, upgrade Node, and run the same <code>npx</code> command again.</p>
|
||||
<p>The selective installer requires Node.js 20.11.0 or newer. Check with <code>node --version</code>, upgrade Node, and run the same <code>npx</code> command again.</p>
|
||||
|
||||
<h3>Unknown skill or profile</h3>
|
||||
<p>Run <code>npx @waybarrios/opencode-power-pack list</code> and copy the lowercase hyphenated ID exactly. Profile names and skill IDs are case-sensitive.</p>
|
||||
@@ -787,6 +829,6 @@ pi remove git:github.com/waybarrios/opencode-power-pack</code></div>
|
||||
</section>
|
||||
</main>
|
||||
</div>
|
||||
<script src="../app.js?v=0.5.0-sandbox-1" defer></script>
|
||||
<script src="../app.js?v=2026-10-19-opencode-2" defer></script>
|
||||
</body>
|
||||
</html>
|
||||
|
||||
+60
-7
@@ -1,10 +1,16 @@
|
||||
# OpenCode Power Pack Installation and Operations
|
||||
|
||||
This guide covers selective npm installation, native host installation, and contained command execution for OpenCode Power Pack v0.5.0.
|
||||
This guide covers selective npm installation, native host installation, and contained command execution for the current main branch and the published npm release.
|
||||
|
||||
## What is available now
|
||||
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
The npm installer still works for its released skills on OpenCode 2. Copying skills does not register the plugin's three named specialists.
|
||||
|
||||
## Choose an installation path
|
||||
|
||||
Use the selective npm installer when Codex, OpenCode, or Pi should load only chosen skills or a curated profile. It requires Node.js 20 or newer and installs into a portable `.agents/skills` directory.
|
||||
Use the selective npm installer when Codex, OpenCode, or Pi should load only chosen skills or a curated profile. It requires Node.js 20.11.0 or newer and installs into a portable `.agents/skills` directory.
|
||||
|
||||
Use the host-native plugin or package installation when you want the full catalog and native activation behavior. Claude Code, Codex, OpenCode, and Pi each use their own host mechanism.
|
||||
|
||||
@@ -25,12 +31,25 @@ npx @waybarrios/opencode-power-pack install code-review --project
|
||||
npx @waybarrios/opencode-power-pack install --profile review --dry-run
|
||||
npx @waybarrios/opencode-power-pack install --profile review
|
||||
|
||||
# Install all 54 skills
|
||||
# Install every skill in the published npm catalog
|
||||
npx @waybarrios/opencode-power-pack install --all
|
||||
```
|
||||
|
||||
The installer resolves required companion skills. Existing skills are not overwritten unless `--force` is supplied. Use `--dry-run` before any forced update.
|
||||
|
||||
## GitHub Packages mirror
|
||||
|
||||
npmjs is the public install source and does not require GitHub authentication. The same released package is also available through GitHub Packages.
|
||||
|
||||
Use a GitHub personal access token (classic) with `read:packages` for the mirror:
|
||||
|
||||
```sh
|
||||
npm login --scope=@waybarrios --auth-type=legacy --registry=https://npm.pkg.github.com
|
||||
npm install --global @waybarrios/opencode-power-pack --registry=https://npm.pkg.github.com
|
||||
```
|
||||
|
||||
The mirror follows release tags. It does not include additions that are only on main.
|
||||
|
||||
## Agent-assisted installation
|
||||
|
||||
A terminal-capable Codex, OpenCode, or Pi agent can install the recommended profile at project scope with a dry-run-first sequence:
|
||||
@@ -63,13 +82,23 @@ codex plugin list --marketplace opencode-power-pack
|
||||
|
||||
Start a new Codex session and invoke a skill such as `$code-review`.
|
||||
|
||||
### OpenCode
|
||||
### OpenCode 1 and 2
|
||||
|
||||
Use Git main for OpenCode 2.0 support and the 55-skill catalog. The v0.5.0 tag contains the earlier V1-only plugin.
|
||||
|
||||
```sh
|
||||
opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git#v0.5.0"
|
||||
# OpenCode 2
|
||||
opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
|
||||
# OpenCode 1
|
||||
opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
```
|
||||
|
||||
Start a new OpenCode session and invoke a skill such as `/code-review`.
|
||||
Restart OpenCode and invoke a skill such as `/code-review` or `/humanizer`. On V2, run `opencode debug agents` to inspect the three specialists. On V1, use `opencode debug skill` and `opencode debug agent code-explorer`.
|
||||
|
||||
V2 enforces the bundled specialists' read-only restrictions when it evaluates permissions, even with global allow rules. A custom system prompt or primary mode replaces the bundled role and keeps host permission defaults. The debug listing can still show the host's base permissions.
|
||||
|
||||
Smoke tests cover OpenCode 1.18.7, 2.0.14, and 2.0.32. They verify all 55 skills and three specialists without model calls.
|
||||
|
||||
### Pi
|
||||
|
||||
@@ -185,9 +214,20 @@ macOS and Linux are supported. WSL2 is expected through the Linux backend but st
|
||||
- `frontend`: frontend implementation plus anti-generic design critique.
|
||||
- `security`: AppSec review, scanning, modeling, validation, and reporting.
|
||||
- `huggingface`: model discovery, inference, training, evaluation, Spaces, and AWS deployment.
|
||||
- `authoring`: skills, MCP servers, and technical-paper summaries.
|
||||
- `authoring`: humanizer, skill-creator, mcp-builder, and paper-summarizer on main. npm 0.5.0 has the three earlier authoring skills.
|
||||
- `project-memory`: durable AGENTS.md and CLAUDE.md maintenance.
|
||||
|
||||
## Humanizer from current source
|
||||
|
||||
Run the selective installer from Git main for a user-level copy:
|
||||
|
||||
```sh
|
||||
npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer --dry-run
|
||||
npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer
|
||||
```
|
||||
|
||||
Use `install --profile authoring` instead for all four authoring skills. For project scope, run the same command from the target repository with `--project`.
|
||||
|
||||
## Update and remove
|
||||
|
||||
Preview an npm-managed skill update, then apply it only if the destination is correct:
|
||||
@@ -199,6 +239,19 @@ npx @waybarrios/opencode-power-pack install code-review --force
|
||||
|
||||
For host-native installations, use that host's plugin or package update and removal commands. Avoid activating the same workflow from both a native plugin and an npm copy at the same scope.
|
||||
|
||||
On OpenCode 2, update or remove with the exact configured package specification:
|
||||
|
||||
```sh
|
||||
opencode plugin update "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
opencode plugin remove "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
|
||||
```
|
||||
|
||||
On V1, reinstall with `plugin --global` to update, or remove its entry from the `plugin` config array. For a local `git+file://` install, pull the clone and then update or reinstall the same specification.
|
||||
|
||||
## Contributor checks
|
||||
|
||||
The behavioral runner uses the OpenCode 2 session API. It rejects partial text from failed or interrupted turns and accepts optional `provider/model#variant` references. Live runs need provider environment credentials; the isolated runtime does not copy the native V2 credential database.
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
- If a new skill is missing, start a new host session and verify the expected `.agents/skills` destination.
|
||||
|
||||
+455
-434
@@ -3,7 +3,7 @@
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1">
|
||||
<title>54 Agent Skills Reference | OpenCode Power Pack</title>
|
||||
<title>55 Agent Skills Reference | OpenCode Power Pack</title>
|
||||
<meta
|
||||
name="description"
|
||||
content="What every OpenCode Power Pack skill does, when to use it, its profiles and dependencies, and the exact npm command to install it for Codex, OpenCode, or Pi."
|
||||
@@ -18,7 +18,7 @@
|
||||
<meta property="og:type" content="article">
|
||||
<meta property="og:locale" content="en_US">
|
||||
<meta property="og:site_name" content="OpenCode Power Pack">
|
||||
<meta property="og:title" content="54 Agent Skills: Complete Reference">
|
||||
<meta property="og:title" content="55 Agent Skills: Complete Reference">
|
||||
<meta
|
||||
property="og:description"
|
||||
content="Detailed purpose, selection guidance, profiles, dependencies, invocation, and individual installation commands for every bundled skill."
|
||||
@@ -28,12 +28,12 @@
|
||||
<meta property="og:image:width" content="1200">
|
||||
<meta property="og:image:height" content="630">
|
||||
<meta property="og:image:type" content="image/png">
|
||||
<meta property="og:image:alt" content="Complete reference for 54 OpenCode Power Pack Agent Skills">
|
||||
<meta property="og:image:alt" content="Complete reference for 55 OpenCode Power Pack Agent Skills">
|
||||
<meta name="twitter:card" content="summary_large_image">
|
||||
<meta name="twitter:title" content="54 Agent Skills: Complete Reference">
|
||||
<meta name="twitter:title" content="55 Agent Skills: Complete Reference">
|
||||
<meta name="twitter:description" content="Purpose, dependencies, invocation, and exact install command for every OpenCode Power Pack skill.">
|
||||
<meta name="twitter:image" content="https://skills.waybarrios.com/assets/social-card.png">
|
||||
<meta name="twitter:image:alt" content="Complete reference for 54 OpenCode Power Pack Agent Skills">
|
||||
<meta name="twitter:image:alt" content="Complete reference for 55 OpenCode Power Pack Agent Skills">
|
||||
<meta name="twitter:image:width" content="1200">
|
||||
<meta name="twitter:image:height" content="630">
|
||||
<link rel="preconnect" href="https://fonts.googleapis.com">
|
||||
@@ -42,434 +42,441 @@
|
||||
href="https://fonts.googleapis.com/css2?family=Barlow+Condensed:wght@600;700;800&family=IBM+Plex+Mono:wght@400;500;600;700&display=swap"
|
||||
rel="stylesheet"
|
||||
>
|
||||
<link rel="stylesheet" href="../../styles.css?v=0.5.0-sandbox-1">
|
||||
<link rel="stylesheet" href="../../styles.css?v=2026-10-19-opencode-2">
|
||||
<script type="application/ld+json">
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@type": "CollectionPage",
|
||||
"name": "OpenCode Power Pack Skill Reference",
|
||||
"description": "Detailed documentation and individual installation commands for 55 Agent Skills.",
|
||||
"url": "https://skills.waybarrios.com/docs/skills/",
|
||||
"isPartOf": {
|
||||
"@type": "WebSite",
|
||||
"name": "OpenCode Power Pack",
|
||||
"url": "https://skills.waybarrios.com/"
|
||||
},
|
||||
"mainEntity": {
|
||||
"@type": "ItemList",
|
||||
"numberOfItems": 55,
|
||||
"itemListElement": [
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@type": "CollectionPage",
|
||||
"name": "OpenCode Power Pack Skill Reference",
|
||||
"description": "Detailed documentation and individual installation commands for 54 Agent Skills.",
|
||||
"url": "https://skills.waybarrios.com/docs/skills/",
|
||||
"isPartOf": {
|
||||
"@type": "WebSite",
|
||||
"name": "OpenCode Power Pack",
|
||||
"url": "https://skills.waybarrios.com/"
|
||||
},
|
||||
"mainEntity": {
|
||||
"@type": "ItemList",
|
||||
"numberOfItems": 54,
|
||||
"itemListElement": [
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agentic-actions-auditor",
|
||||
"name": "agentic-actions-auditor",
|
||||
"description": "Audit GitHub Actions that run AI agents for prompt injection, unsafe interpolation, sandbox gaps, and permissive actor rules. Use for agentic CI workflows, not general application code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agents-md-improver",
|
||||
"name": "agents-md-improver",
|
||||
"description": "Audit and improve project-rules files (AGENTS.md, CLAUDE.md, .agents/instructions, local overrides) so the agent keeps accurate project context. Use when the user asks to check, audit, review, update, improve, or fix their AGENTS.md or CLAUDE.md, mentions \"project rules maintenance\" or \"agent context optimization\", or when the codebase has changed enough that the rules file may be stale. Scans the repository for every rules file, grades each against a quality rubric, outputs a quality report, and applies targeted edits only after user approval."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 3,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agents-md-revise",
|
||||
"name": "agents-md-revise",
|
||||
"description": "Capture learnings from the current session into the project-rules file (AGENTS.md, CLAUDE.md, or local override) so future sessions benefit. Use when the user says \"revise the rules\", \"update AGENTS.md / CLAUDE.md with what we just learned\", \"save this to project memory\", \"remember this for next time\", or at the end of a productive session when valuable context has emerged that is not yet documented. This complements agents-md-improver; improver audits, while this one captures."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 4,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#ai-slop",
|
||||
"name": "ai-slop",
|
||||
"description": "Operational rubric that turns \"don't make AI slop\" into observable properties, severity levels, evidence requirements, and repair actions for interface design. Use as the reference rubric when building or reviewing marketing sites, product interfaces, dashboards, portfolios, or e-commerce pages, especially alongside frontend-design."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 5,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-architect",
|
||||
"name": "code-architect",
|
||||
"description": "Design a feature architecture by analyzing existing codebase patterns and conventions, then provide a comprehensive implementation blueprint with specific files to create or modify, component designs, data flows, and a build sequence. Use this skill when the user asks for an architecture design, an implementation plan for a non-trivial feature, or when dispatched as a sub-task during feature-dev architecture phase."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 6,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-explorer",
|
||||
"name": "code-explorer",
|
||||
"description": "Deeply analyze an existing codebase feature by tracing execution paths, mapping architecture layers, understanding patterns and abstractions, and documenting dependencies. Use this skill when you need to understand how a feature works before modifying or extending it, when dispatched as a sub-task during feature-dev exploration, or when the user asks \"how does X work in this codebase\"."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 7,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-quality",
|
||||
"name": "code-quality",
|
||||
"description": "Agents should invoke this skill for code reviews, linting/formatting setup, maintainability checks, complexity concerns, warning cleanup, coding standards, or quality gates in Rust, TypeScript, Python, shell, and mixed repos."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 8,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-review",
|
||||
"name": "code-review",
|
||||
"description": "Review a pull request or a set of code changes for bugs, logic errors, and project-convention violations using a confidence-filtered, multi-agent process. Use this skill when the user asks to review a PR, audit pending changes, or inspect a diff for problems before merging."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 9,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-reviewer",
|
||||
"name": "code-reviewer",
|
||||
"description": "Review code for bugs, logic errors, security vulnerabilities, code quality issues, and adherence to project conventions, using confidence-based filtering to report only high-priority issues that truly matter. Use this skill when reviewing a small set of changes locally (such as unstaged diff), when dispatched as a sub-task during feature-dev quality review, or when the user wants a critique of a specific file or function."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 10,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#codeql",
|
||||
"name": "codeql",
|
||||
"description": "Run CodeQL database creation and security queries, add data-extension models, or process CodeQL SARIF. Use when CodeQL is explicitly requested; use security-review for a broader manual security review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 11,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#design-patterns",
|
||||
"name": "design-patterns",
|
||||
"description": "Agents should invoke this skill when choosing patterns, designing traits/interfaces/components, deciding abstraction boundaries, evaluating dependency injection/callbacks, or comparing implementation approaches in Rust, TypeScript/React, or Django/Python."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 12,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#differential-review",
|
||||
"name": "differential-review",
|
||||
"description": "Performs security-focused differential review of code changes (PRs, commits, diffs). Adapts analysis depth to codebase size, uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports. Automatically detects and prevents security regressions."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 13,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#feature-dev",
|
||||
"name": "feature-dev",
|
||||
"description": "Guide a feature implementation through a structured seven-phase workflow with deep codebase understanding, clarifying questions, parallel architecture design, and quality review. Use this skill when the user asks to build a new feature, add functionality, or wants a methodical approach to implementation rather than diving straight to code."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 14,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#fp-check",
|
||||
"name": "fp-check",
|
||||
"description": "Systematically verifies suspected security bugs to eliminate false positives. Produces TRUE POSITIVE or FALSE POSITIVE verdicts with documented evidence for each bug."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 15,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#frontend-design",
|
||||
"name": "frontend-design",
|
||||
"description": "Create distinctive, production-grade frontend interfaces with high design quality and accessible markup. Use this skill when the user asks to build or beautify web components, pages, applications, landing pages, dashboards, artifacts, or React/HTML/CSS UI. Generates creative, polished code that avoids generic AI aesthetics, then self-checks it against an objective accessibility and quality rubric."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 16,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cli",
|
||||
"name": "hf-cli",
|
||||
"description": "Use the `hf` CLI for Hub authentication, downloads, uploads, repositories, cache, jobs, buckets, webhooks, and endpoint administration. Use for CLI operations, not model recommendations, paper analysis, training design, or building a Space."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 17,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-aws-context-discovery",
|
||||
"name": "hf-cloud-aws-context-discovery",
|
||||
"description": "Discover the active AWS profile, region, account, and caller identity before SageMaker or AWS work. Use when local AWS context is needed or unspecified; do not use for cloud-agnostic model planning."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 18,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-python-env-setup",
|
||||
"name": "hf-cloud-python-env-setup",
|
||||
"description": "Create an isolated Python environment with a compatible Python version and boto3 for SageMaker or AWS automation. Use before executing AWS Python code; do not use for general Python environment setup."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 19,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-deployment-planner",
|
||||
"name": "hf-cloud-sagemaker-deployment-planner",
|
||||
"description": "Plan and coordinate a model deployment to Amazon SageMaker, including serving stack and real-time versus async inference. Use as the entry point for SageMaker hosting requests, before image, IAM, and endpoint implementation skills."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 20,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-iam-preflight",
|
||||
"name": "hf-cloud-sagemaker-iam-preflight",
|
||||
"description": "Verify or select a SageMaker execution role before creating models, endpoints, or training jobs. Use when a role ARN is missing or IAM access fails; inspect existing roles before proposing role creation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 21,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-production-defaults",
|
||||
"name": "hf-cloud-sagemaker-production-defaults",
|
||||
"description": "Implement a production SageMaker endpoint with autoscaling, CloudWatch alarms, and tags. Use after the serving image and IAM role are known; use the deployment planner first when architecture is undecided."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 22,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-serving-image-selection",
|
||||
"name": "hf-cloud-serving-image-selection",
|
||||
"description": "Select and verify the current region-specific serving container URI for a SageMaker model deployment. Use after the deployment pathway is chosen and before endpoint code; never infer an image URI from memory."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 23,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-mem",
|
||||
"name": "hf-mem",
|
||||
"description": "Hugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 24,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-best",
|
||||
"name": "huggingface-best",
|
||||
"description": "Find and compare recommended Hugging Face models for a task using benchmarks, model size, and device constraints. Use for model selection questions; use huggingface-local-models for GGUF setup and hf-cli for Hub operations."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 25,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-community-evals",
|
||||
"name": "huggingface-community-evals",
|
||||
"description": "Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Use for backend selection, local GPU evals, and choosing between vLLM / Transformers / accelerate. Not for HF Jobs orchestration, model-card PRs, .eval_results publication, or community-evals automation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 26,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-datasets",
|
||||
"name": "huggingface-datasets",
|
||||
"description": "Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 27,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-gradio",
|
||||
"name": "huggingface-gradio",
|
||||
"description": "Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 28,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-llm-trainer",
|
||||
"name": "huggingface-llm-trainer",
|
||||
"description": "Train or fine-tune language models with TRL or Unsloth on Hugging Face Jobs, including SFT, DPO, GRPO, reward models, and GGUF conversion. Use for cloud LLM training; use huggingface-vision-trainer for vision tasks."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 29,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-local-models",
|
||||
"name": "huggingface-local-models",
|
||||
"description": "Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 30,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-lora-space-builder",
|
||||
"name": "huggingface-lora-space-builder",
|
||||
"description": "Build and publish a Gradio Hugging Face Space for a user-provided image or video LoRA, including pipeline selection and model-card settings. Use for LoRA demos; use huggingface-spaces for general Space work."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 31,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-paper-publisher",
|
||||
"name": "huggingface-paper-publisher",
|
||||
"description": "Publish and manage research papers on Hugging Face Hub. Supports creating paper pages, linking papers to models/datasets, claiming authorship, and generating professional markdown-based research articles."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 32,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-papers",
|
||||
"name": "huggingface-papers",
|
||||
"description": "Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata such as authors, linked models/datasets/spaces, Github repo and project page. Use when the user shares a Hugging Face paper page URL, an arXiv URL or ID, or asks to summarize, explain, or analyze an AI research paper."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 33,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-spaces",
|
||||
"name": "huggingface-spaces",
|
||||
"description": "Build, deploy, debug, or maintain a Hugging Face Space using Gradio, Docker, or Static SDKs. Use for general Space hosting and configuration; use huggingface-zerogpu for ZeroGPU runtime constraints and lora-space-builder for LoRA demos."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 34,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-tool-builder",
|
||||
"name": "huggingface-tool-builder",
|
||||
"description": "Use this skill when the user wants to build tool/scripts or achieve a task where using data from the Hugging Face API would help. This is especially useful when chaining or combining API calls or the task will be repeated/automated. This Skill creates a reusable script to fetch, enrich or process data."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 35,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-trackio",
|
||||
"name": "huggingface-trackio",
|
||||
"description": "Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API), firing alerts for training diagnostics, or retrieving/analyzing logged metrics (CLI). Supports real-time dashboard visualization, alerts with webhooks, HF Space syncing, and JSON output for automation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 36,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-vision-trainer",
|
||||
"name": "huggingface-vision-trainer",
|
||||
"description": "Train object-detection, image-classification, or SAM segmentation models on Hugging Face Jobs. Use for vision fine-tuning and evaluation; use huggingface-llm-trainer for language models."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 37,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-zerogpu",
|
||||
"name": "huggingface-zerogpu",
|
||||
"description": "Implement or debug Hugging Face Spaces ZeroGPU code using `@spaces.GPU`, including isolation, state, duration, sizing, and CUDA-wheel constraints. Use for ZeroGPU-specific behavior; use huggingface-spaces for general hosting."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 38,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#insecure-defaults",
|
||||
"name": "insecure-defaults",
|
||||
"description": "Detects fail-open insecure defaults (hardcoded secrets, weak auth, permissive security) that allow apps to run insecurely in production. Use when auditing security, reviewing config management, or analyzing environment variable handling."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 39,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#mcp-builder",
|
||||
"name": "mcp-builder",
|
||||
"description": "Guide the creation of high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when the user wants to build an MCP server to integrate an external API or service, whether in Python (FastMCP) or Node/TypeScript (MCP SDK)."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 40,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#paper-summarizer",
|
||||
"name": "paper-summarizer",
|
||||
"description": "Agents should invoke this skill for academic or technical papers, arXiv/PubMed/IEEE/ACM links, PDFs, methodology review, limitations, practical implications, or extracting findings for engineering decisions."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 41,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#sarif-parsing",
|
||||
"name": "sarif-parsing",
|
||||
"description": "Parses and processes SARIF files from static analysis tools like CodeQL, Semgrep, or other scanners. Triggers on \\\"parse sarif\\\", \\\"read scan results\\\", \\\"aggregate findings\\\", \\\"deduplicate alerts\\\", or \\\"process sarif output\\\". Handles filtering, deduplication, format conversion, and CI/CD integration of SARIF data. Does NOT run scans; use the Semgrep or CodeQL skills for that."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 42,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#security-review",
|
||||
"name": "security-review",
|
||||
"description": "Perform a focused security review of pending git changes to identify high-confidence security vulnerabilities with real exploitation potential. Use this skill when the user asks for a security review, security audit, vulnerability scan, or wants to check pending changes on a branch for security issues before merging. This is NOT a general code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 43,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#security-threat-model",
|
||||
"name": "security-threat-model",
|
||||
"description": "Create a repository-grounded AppSec threat model covering assets, trust boundaries, attackers, abuse paths, and mitigations. Use only for explicit threat-modeling requests, not general architecture or code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 44,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep",
|
||||
"name": "semgrep",
|
||||
"description": "Run Semgrep static analysis across a codebase, optionally using Semgrep Pro for cross-file taint analysis. Use when Semgrep or a static-analysis scan is requested; use security-review for a manual audit."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 45,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep-rule-creator",
|
||||
"name": "semgrep-rule-creator",
|
||||
"description": "Creates custom Semgrep rules for detecting security vulnerabilities, bug patterns, and code patterns. Use when writing Semgrep rules or building custom static analysis detections."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 46,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep-rule-variant-creator",
|
||||
"name": "semgrep-rule-variant-creator",
|
||||
"description": "Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 47,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#sharp-edges",
|
||||
"name": "sharp-edges",
|
||||
"description": "Identify error-prone APIs, dangerous configuration, and footgun designs that enable security mistakes. Use for API ergonomics, misuse resistance, secure defaults, or pit-of-success reviews; use code-review for general defects."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 48,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#skill-creator",
|
||||
"name": "skill-creator",
|
||||
"description": "Create new skills (SKILL.md files), modify and improve existing skills, and design skill descriptions for accurate triggering. Use when the user wants to create a new skill from scratch, edit an existing skill, optimize a skill's description, or convert a workflow they just demonstrated into a reusable skill."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 49,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#supply-chain-risk-auditor",
|
||||
"name": "supply-chain-risk-auditor",
|
||||
"description": "Identifies dependencies at heightened risk of exploitation or takeover. Use when assessing supply chain attack surface, evaluating dependency health, or scoping security engagements."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 50,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#train-sentence-transformers",
|
||||
"name": "train-sentence-transformers",
|
||||
"description": "Train or fine-tune SentenceTransformer bi-encoders, CrossEncoder rerankers, or SparseEncoder models, including losses, negatives, evaluation, distillation, LoRA, and Matryoshka. Use for sentence-transformers training tasks."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 51,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#transformers-js",
|
||||
"name": "transformers-js",
|
||||
"description": "Run Hugging Face models in JavaScript or TypeScript with Transformers.js, WebGPU, or WASM across browser, Node.js, Bun, and Deno. Use for client-side or JS-runtime inference, not Python training."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 52,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#trl-training",
|
||||
"name": "trl-training",
|
||||
"description": "Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 53,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#variant-analysis",
|
||||
"name": "variant-analysis",
|
||||
"description": "Find similar vulnerabilities and bugs across codebases using pattern-based analysis. Use when hunting bug variants, building CodeQL/Semgrep queries, analyzing security vulnerabilities, or performing systematic code audits after finding an initial issue."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 54,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#vuln-report",
|
||||
"name": "vuln-report",
|
||||
"description": "Turn one confirmed security finding into a disclosure-ready GitHub advisory with root cause, proof of concept, impact, and source evidence. Use for reporting an established vulnerability, not discovering or validating one."
|
||||
}
|
||||
]
|
||||
},
|
||||
"breadcrumb": {
|
||||
"@type": "BreadcrumbList",
|
||||
"itemListElement": [
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"name": "OpenCode Power Pack",
|
||||
"item": "https://skills.waybarrios.com/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"name": "Documentation",
|
||||
"item": "https://skills.waybarrios.com/docs/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 3,
|
||||
"name": "Skill reference",
|
||||
"item": "https://skills.waybarrios.com/docs/skills/"
|
||||
}
|
||||
]
|
||||
}
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agentic-actions-auditor",
|
||||
"name": "agentic-actions-auditor",
|
||||
"description": "Audit GitHub Actions that run AI agents for prompt injection, unsafe interpolation, sandbox gaps, and permissive actor rules. Use for agentic CI workflows, not general application code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agents-md-improver",
|
||||
"name": "agents-md-improver",
|
||||
"description": "Audit and improve project-rules files (AGENTS.md, CLAUDE.md, .agents/instructions, local overrides) so the agent keeps accurate project context. Use when the user asks to check, audit, review, update, improve, or fix their AGENTS.md or CLAUDE.md, mentions \"project rules maintenance\" or \"agent context optimization\", or when the codebase has changed enough that the rules file may be stale. Scans the repository for every rules file, grades each against a quality rubric, outputs a quality report, and applies targeted edits only after user approval."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 3,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#agents-md-revise",
|
||||
"name": "agents-md-revise",
|
||||
"description": "Capture learnings from the current session into the project-rules file (AGENTS.md, CLAUDE.md, or local override) so future sessions benefit. Use when the user says \"revise the rules\", \"update AGENTS.md / CLAUDE.md with what we just learned\", \"save this to project memory\", \"remember this for next time\", or at the end of a productive session when valuable context has emerged that is not yet documented. This complements agents-md-improver; improver audits, while this one captures."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 4,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#ai-slop",
|
||||
"name": "ai-slop",
|
||||
"description": "Operational rubric that turns \"don't make AI slop\" into observable properties, severity levels, evidence requirements, and repair actions for interface design. Use as the reference rubric when building or reviewing marketing sites, product interfaces, dashboards, portfolios, or e-commerce pages, especially alongside frontend-design."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 5,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-architect",
|
||||
"name": "code-architect",
|
||||
"description": "Design a feature architecture by analyzing existing codebase patterns and conventions, then provide a comprehensive implementation blueprint with specific files to create or modify, component designs, data flows, and a build sequence. Use this skill when the user asks for an architecture design, an implementation plan for a non-trivial feature, or when dispatched as a sub-task during feature-dev architecture phase."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 6,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-explorer",
|
||||
"name": "code-explorer",
|
||||
"description": "Deeply analyze an existing codebase feature by tracing execution paths, mapping architecture layers, understanding patterns and abstractions, and documenting dependencies. Use this skill when you need to understand how a feature works before modifying or extending it, when dispatched as a sub-task during feature-dev exploration, or when the user asks \"how does X work in this codebase\"."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 7,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-quality",
|
||||
"name": "code-quality",
|
||||
"description": "Agents should invoke this skill for code reviews, linting/formatting setup, maintainability checks, complexity concerns, warning cleanup, coding standards, or quality gates in Rust, TypeScript, Python, shell, and mixed repos."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 8,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-review",
|
||||
"name": "code-review",
|
||||
"description": "Review a pull request or a set of code changes for bugs, logic errors, and project-convention violations using a confidence-filtered, multi-agent process. Use this skill when the user asks to review a PR, audit pending changes, or inspect a diff for problems before merging."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 9,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#code-reviewer",
|
||||
"name": "code-reviewer",
|
||||
"description": "Review code for bugs, logic errors, security vulnerabilities, code quality issues, and adherence to project conventions, using confidence-based filtering to report only high-priority issues that truly matter. Use this skill when reviewing a small set of changes locally (such as unstaged diff), when dispatched as a sub-task during feature-dev quality review, or when the user wants a critique of a specific file or function."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 10,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#codeql",
|
||||
"name": "codeql",
|
||||
"description": "Run CodeQL database creation and security queries, add data-extension models, or process CodeQL SARIF. Use when CodeQL is explicitly requested; use security-review for a broader manual security review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 11,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#design-patterns",
|
||||
"name": "design-patterns",
|
||||
"description": "Agents should invoke this skill when choosing patterns, designing traits/interfaces/components, deciding abstraction boundaries, evaluating dependency injection/callbacks, or comparing implementation approaches in Rust, TypeScript/React, or Django/Python."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 12,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#differential-review",
|
||||
"name": "differential-review",
|
||||
"description": "Performs security-focused differential review of code changes (PRs, commits, diffs). Adapts analysis depth to codebase size, uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports. Automatically detects and prevents security regressions."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 13,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#feature-dev",
|
||||
"name": "feature-dev",
|
||||
"description": "Guide a feature implementation through a structured seven-phase workflow with deep codebase understanding, clarifying questions, parallel architecture design, and quality review. Use this skill when the user asks to build a new feature, add functionality, or wants a methodical approach to implementation rather than diving straight to code."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 14,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#fp-check",
|
||||
"name": "fp-check",
|
||||
"description": "Systematically verifies suspected security bugs to eliminate false positives. Produces TRUE POSITIVE or FALSE POSITIVE verdicts with documented evidence for each bug."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 15,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#frontend-design",
|
||||
"name": "frontend-design",
|
||||
"description": "Create distinctive, production-grade frontend interfaces with high design quality and accessible markup. Use this skill when the user asks to build or beautify web components, pages, applications, landing pages, dashboards, artifacts, or React/HTML/CSS UI. Generates creative, polished code that avoids generic AI aesthetics, then self-checks it against an objective accessibility and quality rubric."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 16,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cli",
|
||||
"name": "hf-cli",
|
||||
"description": "Use the `hf` CLI for Hub authentication, downloads, uploads, repositories, cache, jobs, buckets, webhooks, and endpoint administration. Use for CLI operations, not model recommendations, paper analysis, training design, or building a Space."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 17,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-aws-context-discovery",
|
||||
"name": "hf-cloud-aws-context-discovery",
|
||||
"description": "Discover the active AWS profile, region, account, and caller identity before SageMaker or AWS work. Use when local AWS context is needed or unspecified; do not use for cloud-agnostic model planning."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 18,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-python-env-setup",
|
||||
"name": "hf-cloud-python-env-setup",
|
||||
"description": "Create an isolated Python environment with a compatible Python version and boto3 for SageMaker or AWS automation. Use before executing AWS Python code; do not use for general Python environment setup."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 19,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-deployment-planner",
|
||||
"name": "hf-cloud-sagemaker-deployment-planner",
|
||||
"description": "Plan and coordinate a model deployment to Amazon SageMaker, including serving stack and real-time versus async inference. Use as the entry point for SageMaker hosting requests, before image, IAM, and endpoint implementation skills."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 20,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-iam-preflight",
|
||||
"name": "hf-cloud-sagemaker-iam-preflight",
|
||||
"description": "Verify or select a SageMaker execution role before creating models, endpoints, or training jobs. Use when a role ARN is missing or IAM access fails; inspect existing roles before proposing role creation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 21,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-sagemaker-production-defaults",
|
||||
"name": "hf-cloud-sagemaker-production-defaults",
|
||||
"description": "Implement a production SageMaker endpoint with autoscaling, CloudWatch alarms, and tags. Use after the serving image and IAM role are known; use the deployment planner first when architecture is undecided."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 22,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-cloud-serving-image-selection",
|
||||
"name": "hf-cloud-serving-image-selection",
|
||||
"description": "Select and verify the current region-specific serving container URI for a SageMaker model deployment. Use after the deployment pathway is chosen and before endpoint code; never infer an image URI from memory."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 23,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#hf-mem",
|
||||
"name": "hf-mem",
|
||||
"description": "Hugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 24,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-best",
|
||||
"name": "huggingface-best",
|
||||
"description": "Find and compare recommended Hugging Face models for a task using benchmarks, model size, and device constraints. Use for model selection questions; use huggingface-local-models for GGUF setup and hf-cli for Hub operations."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 25,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-community-evals",
|
||||
"name": "huggingface-community-evals",
|
||||
"description": "Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Use for backend selection, local GPU evals, and choosing between vLLM / Transformers / accelerate. Not for HF Jobs orchestration, model-card PRs, .eval_results publication, or community-evals automation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 26,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-datasets",
|
||||
"name": "huggingface-datasets",
|
||||
"description": "Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 27,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-gradio",
|
||||
"name": "huggingface-gradio",
|
||||
"description": "Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 28,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-llm-trainer",
|
||||
"name": "huggingface-llm-trainer",
|
||||
"description": "Train or fine-tune language models with TRL or Unsloth on Hugging Face Jobs, including SFT, DPO, GRPO, reward models, and GGUF conversion. Use for cloud LLM training; use huggingface-vision-trainer for vision tasks."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 29,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-local-models",
|
||||
"name": "huggingface-local-models",
|
||||
"description": "Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 30,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-lora-space-builder",
|
||||
"name": "huggingface-lora-space-builder",
|
||||
"description": "Build and publish a Gradio Hugging Face Space for a user-provided image or video LoRA, including pipeline selection and model-card settings. Use for LoRA demos; use huggingface-spaces for general Space work."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 31,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-paper-publisher",
|
||||
"name": "huggingface-paper-publisher",
|
||||
"description": "Publish and manage research papers on Hugging Face Hub. Supports creating paper pages, linking papers to models/datasets, claiming authorship, and generating professional markdown-based research articles."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 32,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-papers",
|
||||
"name": "huggingface-papers",
|
||||
"description": "Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata such as authors, linked models/datasets/spaces, Github repo and project page. Use when the user shares a Hugging Face paper page URL, an arXiv URL or ID, or asks to summarize, explain, or analyze an AI research paper."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 33,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-spaces",
|
||||
"name": "huggingface-spaces",
|
||||
"description": "Build, deploy, debug, or maintain a Hugging Face Space using Gradio, Docker, or Static SDKs. Use for general Space hosting and configuration; use huggingface-zerogpu for ZeroGPU runtime constraints and lora-space-builder for LoRA demos."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 34,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-tool-builder",
|
||||
"name": "huggingface-tool-builder",
|
||||
"description": "Use this skill when the user wants to build tool/scripts or achieve a task where using data from the Hugging Face API would help. This is especially useful when chaining or combining API calls or the task will be repeated/automated. This Skill creates a reusable script to fetch, enrich or process data."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 35,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-trackio",
|
||||
"name": "huggingface-trackio",
|
||||
"description": "Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API), firing alerts for training diagnostics, or retrieving/analyzing logged metrics (CLI). Supports real-time dashboard visualization, alerts with webhooks, HF Space syncing, and JSON output for automation."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 36,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-vision-trainer",
|
||||
"name": "huggingface-vision-trainer",
|
||||
"description": "Train object-detection, image-classification, or SAM segmentation models on Hugging Face Jobs. Use for vision fine-tuning and evaluation; use huggingface-llm-trainer for language models."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 37,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#huggingface-zerogpu",
|
||||
"name": "huggingface-zerogpu",
|
||||
"description": "Implement or debug Hugging Face Spaces ZeroGPU code using `@spaces.GPU`, including isolation, state, duration, sizing, and CUDA-wheel constraints. Use for ZeroGPU-specific behavior; use huggingface-spaces for general hosting."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 38,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#humanizer",
|
||||
"name": "humanizer",
|
||||
"description": "Rewrite English and Spanish copy in a direct, natural voice. Keep facts, names, numbers, URLs, and code intact."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 39,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#insecure-defaults",
|
||||
"name": "insecure-defaults",
|
||||
"description": "Detects fail-open insecure defaults (hardcoded secrets, weak auth, permissive security) that allow apps to run insecurely in production. Use when auditing security, reviewing config management, or analyzing environment variable handling."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 40,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#mcp-builder",
|
||||
"name": "mcp-builder",
|
||||
"description": "Guide the creation of high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when the user wants to build an MCP server to integrate an external API or service, whether in Python (FastMCP) or Node/TypeScript (MCP SDK)."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 41,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#paper-summarizer",
|
||||
"name": "paper-summarizer",
|
||||
"description": "Agents should invoke this skill for academic or technical papers, arXiv/PubMed/IEEE/ACM links, PDFs, methodology review, limitations, practical implications, or extracting findings for engineering decisions."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 42,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#sarif-parsing",
|
||||
"name": "sarif-parsing",
|
||||
"description": "Parses and processes SARIF files from static analysis tools like CodeQL, Semgrep, or other scanners. Triggers on \\\"parse sarif\\\", \\\"read scan results\\\", \\\"aggregate findings\\\", \\\"deduplicate alerts\\\", or \\\"process sarif output\\\". Handles filtering, deduplication, format conversion, and CI/CD integration of SARIF data. Does NOT run scans; use the Semgrep or CodeQL skills for that."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 43,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#security-review",
|
||||
"name": "security-review",
|
||||
"description": "Perform a focused security review of pending git changes to identify high-confidence security vulnerabilities with real exploitation potential. Use this skill when the user asks for a security review, security audit, vulnerability scan, or wants to check pending changes on a branch for security issues before merging. This is NOT a general code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 44,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#security-threat-model",
|
||||
"name": "security-threat-model",
|
||||
"description": "Create a repository-grounded AppSec threat model covering assets, trust boundaries, attackers, abuse paths, and mitigations. Use only for explicit threat-modeling requests, not general architecture or code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 45,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep",
|
||||
"name": "semgrep",
|
||||
"description": "Run Semgrep static analysis across a codebase, optionally using Semgrep Pro for cross-file taint analysis. Use when Semgrep or a static-analysis scan is requested; use security-review for a manual audit."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 46,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep-rule-creator",
|
||||
"name": "semgrep-rule-creator",
|
||||
"description": "Creates custom Semgrep rules for detecting security vulnerabilities, bug patterns, and code patterns. Use when writing Semgrep rules or building custom static analysis detections."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 47,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#semgrep-rule-variant-creator",
|
||||
"name": "semgrep-rule-variant-creator",
|
||||
"description": "Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 48,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#sharp-edges",
|
||||
"name": "sharp-edges",
|
||||
"description": "Identify error-prone APIs, dangerous configuration, and footgun designs that enable security mistakes. Use for API ergonomics, misuse resistance, secure defaults, or pit-of-success reviews; use code-review for general defects."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 49,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#skill-creator",
|
||||
"name": "skill-creator",
|
||||
"description": "Create new skills (SKILL.md files), modify and improve existing skills, and design skill descriptions for accurate triggering. Use when the user wants to create a new skill from scratch, edit an existing skill, optimize a skill's description, or convert a workflow they just demonstrated into a reusable skill."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 50,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#supply-chain-risk-auditor",
|
||||
"name": "supply-chain-risk-auditor",
|
||||
"description": "Identifies dependencies at heightened risk of exploitation or takeover. Use when assessing supply chain attack surface, evaluating dependency health, or scoping security engagements."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 51,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#train-sentence-transformers",
|
||||
"name": "train-sentence-transformers",
|
||||
"description": "Train or fine-tune SentenceTransformer bi-encoders, CrossEncoder rerankers, or SparseEncoder models, including losses, negatives, evaluation, distillation, LoRA, and Matryoshka. Use for sentence-transformers training tasks."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 52,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#transformers-js",
|
||||
"name": "transformers-js",
|
||||
"description": "Run Hugging Face models in JavaScript or TypeScript with Transformers.js, WebGPU, or WASM across browser, Node.js, Bun, and Deno. Use for client-side or JS-runtime inference, not Python training."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 53,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#trl-training",
|
||||
"name": "trl-training",
|
||||
"description": "Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 54,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#variant-analysis",
|
||||
"name": "variant-analysis",
|
||||
"description": "Find similar vulnerabilities and bugs across codebases using pattern-based analysis. Use when hunting bug variants, building CodeQL/Semgrep queries, analyzing security vulnerabilities, or performing systematic code audits after finding an initial issue."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 55,
|
||||
"url": "https://skills.waybarrios.com/docs/skills/#vuln-report",
|
||||
"name": "vuln-report",
|
||||
"description": "Turn one confirmed security finding into a disclosure-ready GitHub advisory with root cause, proof of concept, impact, and source evidence. Use for reporting an established vulnerability, not discovering or validating one."
|
||||
}
|
||||
]
|
||||
},
|
||||
"breadcrumb": {
|
||||
"@type": "BreadcrumbList",
|
||||
"itemListElement": [
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"name": "OpenCode Power Pack",
|
||||
"item": "https://skills.waybarrios.com/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"name": "Documentation",
|
||||
"item": "https://skills.waybarrios.com/docs/"
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 3,
|
||||
"name": "Skill reference",
|
||||
"item": "https://skills.waybarrios.com/docs/skills/"
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
</script>
|
||||
</head>
|
||||
<body class="docs-body">
|
||||
<a class="skip-link" href="#skills-main">Skip to skill reference</a>
|
||||
<div class="status-rail" aria-hidden="true">
|
||||
<span>SKILL REFERENCE // v0.5.0</span>
|
||||
<span>54 DOCUMENTED</span>
|
||||
<span>SKILL REFERENCE // MAIN</span>
|
||||
<span>55 DOCUMENTED</span>
|
||||
<span>INDIVIDUAL INSTALLS</span>
|
||||
</div>
|
||||
|
||||
@@ -484,7 +491,7 @@
|
||||
<nav aria-label="Skill reference sections">
|
||||
<a href="#install-one">Install one skill</a>
|
||||
<a href="#invoke">Invoke by host</a>
|
||||
<a href="#core">Core skills (14)</a>
|
||||
<a href="#core">Core skills (15)</a>
|
||||
<a href="#security">Security skills (15)</a>
|
||||
<a href="#ml">ML + Hugging Face (25)</a>
|
||||
</nav>
|
||||
@@ -508,10 +515,12 @@
|
||||
</p>
|
||||
</header>
|
||||
|
||||
<p class="note">The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer. <a href="../#opencode-2">Read the compatibility guide.</a></p>
|
||||
|
||||
<section class="docs-section skill-install-guide" id="install-one">
|
||||
<h2>Install one skill</h2>
|
||||
<p>
|
||||
Use npm when you want a selective set for Codex, OpenCode, or Pi. Node.js 20 or newer is
|
||||
Use npm when you want a selective set for Codex, OpenCode, or Pi. Node.js 20.11.0 or newer is
|
||||
required. The default user destination is <code>~/.agents/skills</code>.
|
||||
</p>
|
||||
<div class="docs-code"><code>npx @waybarrios/opencode-power-pack install <skill-id>
|
||||
@@ -534,7 +543,7 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
</div>
|
||||
<p class="note">
|
||||
Native Claude Code, Codex, OpenCode, and Pi plugin/package commands install the full catalog.
|
||||
See the <a href="../#hosts">host installation guide</a> when you want all 54 skills.
|
||||
See the <a href="../#hosts">host installation guide</a> when you want all 55 skills.
|
||||
</p>
|
||||
</section>
|
||||
|
||||
@@ -556,7 +565,7 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
<div class="skill-search">
|
||||
<label for="skill-filter">Filter by name, task, profile, or companion</label>
|
||||
<input id="skill-filter" type="search" placeholder="Try: PR review, AWS, dataset, frontend">
|
||||
<p id="filter-status" aria-live="polite">Showing all 54 skills</p>
|
||||
<p id="filter-status" aria-live="polite">Showing all 55 skills</p>
|
||||
</div>
|
||||
</section>
|
||||
<p class="skill-filter-empty" id="skill-filter-empty" hidden>
|
||||
@@ -568,7 +577,7 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
<div class="skill-reference-group-heading">
|
||||
<span>CORE</span>
|
||||
<h2 id="core-title">Build, review, and author</h2>
|
||||
<b>14</b>
|
||||
<b>15</b>
|
||||
</div>
|
||||
<div class="skill-reference-list">
|
||||
<article class="skill-reference-card" id="agents-md-improver" data-skill="agents-md-improver audit and improve project-rules files (agents.md, claude.md, .agents/instructions, local overrides) so the agent keeps accurate project context. use when the user asks to check, audit, review, update, improve, or fix their agents.md or claude.md, mentions "project rules maintenance" or "agent context optimization", or when the codebase has changed enough that the rules file may be stale. scans the repository for every rules file, grades each against a quality rubric, outputs a quality report, and applies targeted edits only after user approval. core recommended project-memory">
|
||||
@@ -934,9 +943,21 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
<a href="../#hosts">Host setup</a>
|
||||
</footer>
|
||||
</article>
|
||||
<article class="skill-reference-card" id="humanizer" data-skill="humanizer writing copy editing english spanish authoring core">
|
||||
<header><span>CORE / 12</span><h3><a href="#humanizer">humanizer</a></h3></header>
|
||||
<div class="skill-reference-purpose"><span>What it does / when to use</span><p class="skill-reference-description">Rewrite English and Spanish copy in a direct, natural voice. Keep facts, names, numbers, URLs, and code intact.</p></div>
|
||||
<div class="skill-reference-install"><span>Install from Git main</span><code>npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer</code><button class="small-copy" type="button" data-copy="npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer">Copy command</button></div>
|
||||
<p>This runs the installer from Git main and installs into <code>~/.agents/skills</code>. Add <code>--dry-run</code> to preview, or <code>--project</code> for the current repository. npm 0.5.0 does not include humanizer.</p>
|
||||
<dl class="skill-reference-meta">
|
||||
<div><dt>Included by profiles</dt><dd><code>authoring</code> on main, alongside mcp-builder, paper-summarizer, and skill-creator.</dd></div>
|
||||
<div><dt>Automatic companions</dt><dd>None</dd></div>
|
||||
<div><dt>Invoke after installation</dt><dd><code>$humanizer</code> in Codex, <code>/humanizer</code> in OpenCode, or <code>/opencode-power-pack:humanizer</code> in Claude Code.</dd></div>
|
||||
</dl>
|
||||
<footer><a href="https://github.com/waybarrios/opencode-power-pack/tree/main/skills/humanizer">Read the complete workflow <span aria-hidden="true">↗</span></a><a href="../#hosts">Host setup</a></footer>
|
||||
</article>
|
||||
<article class="skill-reference-card" id="mcp-builder" data-skill="mcp-builder guide the creation of high-quality mcp (model context protocol) servers that enable llms to interact with external services through well-designed tools. use when the user wants to build an mcp server to integrate an external api or service, whether in python (fastmcp) or node/typescript (mcp sdk). core recommended authoring">
|
||||
<header>
|
||||
<span>CORE / 12</span>
|
||||
<span>CORE / 13</span>
|
||||
<h3><a href="#mcp-builder">mcp-builder</a></h3>
|
||||
</header>
|
||||
<div class="skill-reference-purpose">
|
||||
@@ -969,7 +990,7 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
</article>
|
||||
<article class="skill-reference-card" id="paper-summarizer" data-skill="paper-summarizer agents should invoke this skill for academic or technical papers, arxiv/pubmed/ieee/acm links, pdfs, methodology review, limitations, practical implications, or extracting findings for engineering decisions. core authoring">
|
||||
<header>
|
||||
<span>CORE / 13</span>
|
||||
<span>CORE / 14</span>
|
||||
<h3><a href="#paper-summarizer">paper-summarizer</a></h3>
|
||||
</header>
|
||||
<div class="skill-reference-purpose">
|
||||
@@ -1002,7 +1023,7 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
</article>
|
||||
<article class="skill-reference-card" id="skill-creator" data-skill="skill-creator create new skills (skill.md files), modify and improve existing skills, and design skill descriptions for accurate triggering. use when the user wants to create a new skill from scratch, edit an existing skill, optimize a skill's description, or convert a workflow they just demonstrated into a reusable skill. core recommended authoring">
|
||||
<header>
|
||||
<span>CORE / 14</span>
|
||||
<span>CORE / 15</span>
|
||||
<h3><a href="#skill-creator">skill-creator</a></h3>
|
||||
</header>
|
||||
<div class="skill-reference-purpose">
|
||||
@@ -2377,6 +2398,6 @@ npx @waybarrios/opencode-power-pack install code-review</code></div>
|
||||
</section>
|
||||
</main>
|
||||
</div>
|
||||
<script src="../../app.js?v=0.5.0-sandbox-1" defer></script>
|
||||
<script src="../../app.js?v=2026-10-19-opencode-2" defer></script>
|
||||
</body>
|
||||
</html>
|
||||
|
||||
+18
-3
@@ -1,14 +1,16 @@
|
||||
# OpenCode Power Pack: Complete Skill Reference
|
||||
|
||||
> Purpose, audience, profile membership, dependencies, install command, invocation, and source link for all 54 skills in v0.5.0.
|
||||
> Purpose, audience, profile membership, dependencies, install command, invocation, and source link for all 55 skills on main.
|
||||
|
||||
Canonical HTML browser: https://skills.waybarrios.com/docs/skills/
|
||||
|
||||
Install any skill at user scope with `npx @waybarrios/opencode-power-pack install <skill-id>`. Add `--project` for repository-local scope or `--dry-run` to preview the resolved changes.
|
||||
Install released skills at user scope with `npx @waybarrios/opencode-power-pack install <skill-id>`. Add `--project` for repository-local scope or `--dry-run` to preview the resolved changes.
|
||||
|
||||
Required companion skills are installed automatically. Invoke an installed workflow as `$<skill-id>` in Codex, `/<skill-id>` in OpenCode, or `/opencode-power-pack:<skill-id>` in Claude Code.
|
||||
|
||||
## Core and authoring skills (14)
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
## Core and authoring skills (15)
|
||||
|
||||
### agents-md-improver
|
||||
|
||||
@@ -153,6 +155,19 @@ Create distinctive, production-grade frontend interfaces with high design qualit
|
||||
- HTML details: https://skills.waybarrios.com/docs/skills/#frontend-design
|
||||
- Source: https://github.com/waybarrios/opencode-power-pack/tree/main/skills/frontend-design
|
||||
|
||||
### humanizer
|
||||
|
||||
Rewrite English and Spanish copy in a direct, natural voice. Keep facts, names, numbers, URLs, and code intact.
|
||||
|
||||
- Profiles: `authoring` on main
|
||||
- Automatic companions: None
|
||||
- Install from source: `npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer`
|
||||
- Codex: `$humanizer`
|
||||
- OpenCode 1 and 2: `/humanizer`
|
||||
- Claude Code: `/opencode-power-pack:humanizer`
|
||||
- HTML details: https://skills.waybarrios.com/docs/skills/#humanizer
|
||||
- Source: https://github.com/waybarrios/opencode-power-pack/tree/main/skills/humanizer
|
||||
|
||||
### mcp-builder
|
||||
|
||||
Guide the creation of high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when the user wants to build an MCP server to integrate an external API or service, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
|
||||
|
||||
+234
-143
@@ -3,10 +3,10 @@
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1">
|
||||
<title>OpenCode Power Pack | 54 Skills for AI Coding Agents</title>
|
||||
<title>OpenCode Power Pack | 55 Skills for AI Coding Agents</title>
|
||||
<meta
|
||||
name="description"
|
||||
content="Install 54 production-grade Agent Skills for Claude Code, Codex, OpenCode, and Pi, with an opt-in native sandbox runner for contained commands."
|
||||
content="Use 55 Agent Skills from main for Claude Code, Codex, OpenCode, and Pi, with an opt-in native sandbox runner for contained commands."
|
||||
>
|
||||
<meta name="author" content="Wayner Barrios">
|
||||
<meta name="robots" content="index, follow, max-image-preview:large, max-snippet:-1">
|
||||
@@ -21,7 +21,7 @@
|
||||
<meta property="og:type" content="website">
|
||||
<meta property="og:locale" content="en_US">
|
||||
<meta property="og:site_name" content="OpenCode Power Pack">
|
||||
<meta property="og:title" content="54 Skills. Four Coding Agents. One Power Pack.">
|
||||
<meta property="og:title" content="55 Skills. Four Coding Agents. One Power Pack.">
|
||||
<meta
|
||||
property="og:description"
|
||||
content="Portable code review, security, feature development, frontend, MCP, and ML workflows for Claude Code, Codex, OpenCode, and Pi."
|
||||
@@ -36,14 +36,14 @@
|
||||
<meta property="og:image:type" content="image/png">
|
||||
<meta
|
||||
property="og:image:alt"
|
||||
content="OpenCode Power Pack, 54 skills for Claude Code, Codex, OpenCode, and Pi"
|
||||
content="OpenCode Power Pack, 55 skills for Claude Code, Codex, OpenCode, and Pi"
|
||||
>
|
||||
|
||||
<meta name="twitter:card" content="summary_large_image">
|
||||
<meta name="twitter:title" content="54 Skills. Four Coding Agents. One Power Pack.">
|
||||
<meta name="twitter:title" content="55 Skills. Four Coding Agents. One Power Pack.">
|
||||
<meta
|
||||
name="twitter:description"
|
||||
content="Production-grade Agent Skills for code review, security, feature development, frontend, MCP, and ML workflows."
|
||||
content="Agent Skills for code review, security, feature development, frontend, MCP, and ML workflows."
|
||||
>
|
||||
<meta
|
||||
name="twitter:image"
|
||||
@@ -51,7 +51,7 @@
|
||||
>
|
||||
<meta
|
||||
name="twitter:image:alt"
|
||||
content="OpenCode Power Pack, 54 skills for Claude Code, Codex, OpenCode, and Pi"
|
||||
content="OpenCode Power Pack, 55 skills for Claude Code, Codex, OpenCode, and Pi"
|
||||
>
|
||||
<meta name="twitter:image:width" content="1200">
|
||||
<meta name="twitter:image:height" content="630">
|
||||
@@ -62,118 +62,191 @@
|
||||
href="https://fonts.googleapis.com/css2?family=Barlow+Condensed:wght@600;700;800&family=IBM+Plex+Mono:wght@400;500;600;700&display=swap"
|
||||
rel="stylesheet"
|
||||
>
|
||||
<link rel="stylesheet" href="styles.css?v=0.5.0-sandbox-1">
|
||||
<link rel="stylesheet" href="styles.css?v=2026-10-19-opencode-2">
|
||||
|
||||
<script type="application/ld+json">
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@graph": [
|
||||
{
|
||||
"@type": "WebSite",
|
||||
"@id": "https://skills.waybarrios.com/#website",
|
||||
"url": "https://skills.waybarrios.com/",
|
||||
"name": "OpenCode Power Pack",
|
||||
"alternateName": "opencode-power-pack",
|
||||
"description": "Fifty-four portable software-development workflows for Claude Code, Codex, OpenCode, and Pi.",
|
||||
"inLanguage": "en",
|
||||
"publisher": { "@id": "https://github.com/waybarrios#person" }
|
||||
},
|
||||
{
|
||||
"@type": "SoftwareSourceCode",
|
||||
"@id": "https://skills.waybarrios.com/#software",
|
||||
"name": "OpenCode Power Pack",
|
||||
"alternateName": "@waybarrios/opencode-power-pack",
|
||||
"description": "A curated collection of 54 Agent Skills for code review, security auditing, feature development, frontend design, MCP authoring, and machine-learning workflows.",
|
||||
"codeRepository": "https://github.com/waybarrios/opencode-power-pack",
|
||||
"downloadUrl": "https://www.npmjs.com/package/@waybarrios/opencode-power-pack",
|
||||
"softwareVersion": "0.5.0",
|
||||
"runtimePlatform": ["Claude Code", "Codex", "OpenCode", "Pi"],
|
||||
"programmingLanguage": ["Markdown", "JavaScript", "Python"],
|
||||
"license": [
|
||||
"https://spdx.org/licenses/MIT.html",
|
||||
"https://spdx.org/licenses/Apache-2.0.html"
|
||||
],
|
||||
"sameAs": [
|
||||
"https://github.com/waybarrios/opencode-power-pack",
|
||||
"https://www.npmjs.com/package/@waybarrios/opencode-power-pack",
|
||||
"https://github.com/users/waybarrios/packages/npm/package/opencode-power-pack"
|
||||
],
|
||||
"isAccessibleForFree": true,
|
||||
"author": { "@id": "https://github.com/waybarrios#person" },
|
||||
"mainEntityOfPage": { "@id": "https://skills.waybarrios.com/#website" },
|
||||
"keywords": "Agent Skills, Claude Code, Codex, OpenCode, Pi, code review, security review, feature development, MCP, Hugging Face"
|
||||
},
|
||||
{
|
||||
"@type": "Person",
|
||||
"@id": "https://github.com/waybarrios#person",
|
||||
"name": "Wayner Barrios",
|
||||
"url": "https://github.com/waybarrios",
|
||||
"sameAs": ["https://github.com/waybarrios"]
|
||||
},
|
||||
{
|
||||
"@type": "FAQPage",
|
||||
"@id": "https://skills.waybarrios.com/#faq",
|
||||
"mainEntity": [
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "What is an Agent Skill?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "An Agent Skill is a reusable workflow centered on a SKILL.md file. It provides durable instructions, references, scripts, and decision rules for a specific job."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Does the same skill work in Claude Code, Codex, OpenCode, and Pi?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "The stable workflow IDs are shared, while packaging and activation follow each host's native model."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "What is the difference between code-review and code-reviewer?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "code-review is a comprehensive, merge-oriented review for a pull request or complete change set. code-reviewer is a focused implementation review for small changes and feature-development handoffs."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Can I install only one skill?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "Yes. Run npx @waybarrios/opencode-power-pack install followed by the skill name. The installer also includes required companion skills."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Does the sandbox work with Codex and OpenCode?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "Yes. Codex and OpenCode can explicitly run commands through the shared sandbox CLI on supported Linux and macOS hosts. Automatic routing and whole-agent isolation are not yet implemented."
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"@type": "ItemList",
|
||||
"@id": "https://skills.waybarrios.com/#profiles",
|
||||
"name": "OpenCode Power Pack installable profiles",
|
||||
"numberOfItems": 8,
|
||||
"itemListElement": [
|
||||
{ "@type": "ListItem", "position": 1, "name": "recommended", "url": "https://skills.waybarrios.com/#profiles", "description": "Balanced everyday software-engineering workflows." },
|
||||
{ "@type": "ListItem", "position": 2, "name": "review", "url": "https://skills.waybarrios.com/#profiles", "description": "Comprehensive, focused, quality, and differential code review." },
|
||||
{ "@type": "ListItem", "position": 3, "name": "feature-dev", "url": "https://skills.waybarrios.com/#profiles", "description": "Structured feature exploration, architecture, implementation, and review." },
|
||||
{ "@type": "ListItem", "position": 4, "name": "frontend", "url": "https://skills.waybarrios.com/#profiles", "description": "Frontend implementation and anti-generic design critique." },
|
||||
{ "@type": "ListItem", "position": 5, "name": "security", "url": "https://skills.waybarrios.com/#profiles", "description": "Application-security review, scanning, validation, threat modeling, and reporting." },
|
||||
{ "@type": "ListItem", "position": 6, "name": "huggingface", "url": "https://skills.waybarrios.com/#profiles", "description": "Model discovery, inference, training, evaluation, Spaces, and AWS deployment." },
|
||||
{ "@type": "ListItem", "position": 7, "name": "authoring", "url": "https://skills.waybarrios.com/#profiles", "description": "Agent Skill, MCP server, and technical-research authoring." },
|
||||
{ "@type": "ListItem", "position": 8, "name": "project-memory", "url": "https://skills.waybarrios.com/#profiles", "description": "Durable AGENTS.md and CLAUDE.md project context." }
|
||||
]
|
||||
}
|
||||
]
|
||||
{
|
||||
"@context": "https://schema.org",
|
||||
"@graph": [
|
||||
{
|
||||
"@type": "WebSite",
|
||||
"@id": "https://skills.waybarrios.com/#website",
|
||||
"url": "https://skills.waybarrios.com/",
|
||||
"name": "OpenCode Power Pack",
|
||||
"alternateName": "opencode-power-pack",
|
||||
"description": "Fifty-five portable software-development workflows for Claude Code, Codex, OpenCode, and Pi.",
|
||||
"inLanguage": "en",
|
||||
"publisher": {
|
||||
"@id": "https://github.com/waybarrios#person"
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "SoftwareSourceCode",
|
||||
"@id": "https://skills.waybarrios.com/#software",
|
||||
"name": "OpenCode Power Pack",
|
||||
"alternateName": "@waybarrios/opencode-power-pack",
|
||||
"description": "A curated collection of 55 Agent Skills for code review, security auditing, feature development, frontend design, MCP authoring, and machine-learning workflows.",
|
||||
"codeRepository": "https://github.com/waybarrios/opencode-power-pack",
|
||||
"downloadUrl": "https://www.npmjs.com/package/@waybarrios/opencode-power-pack",
|
||||
"softwareVersion": "main",
|
||||
"runtimePlatform": [
|
||||
"Claude Code",
|
||||
"Codex",
|
||||
"OpenCode 1 and 2",
|
||||
"Pi"
|
||||
],
|
||||
"programmingLanguage": [
|
||||
"Markdown",
|
||||
"JavaScript",
|
||||
"Python"
|
||||
],
|
||||
"license": [
|
||||
"https://spdx.org/licenses/MIT.html",
|
||||
"https://spdx.org/licenses/Apache-2.0.html"
|
||||
],
|
||||
"sameAs": [
|
||||
"https://github.com/waybarrios/opencode-power-pack",
|
||||
"https://www.npmjs.com/package/@waybarrios/opencode-power-pack",
|
||||
"https://github.com/users/waybarrios/packages/npm/package/opencode-power-pack"
|
||||
],
|
||||
"isAccessibleForFree": true,
|
||||
"author": {
|
||||
"@id": "https://github.com/waybarrios#person"
|
||||
},
|
||||
"mainEntityOfPage": {
|
||||
"@id": "https://skills.waybarrios.com/#website"
|
||||
},
|
||||
"keywords": "Agent Skills, Claude Code, Codex, OpenCode, Pi, code review, security review, feature development, MCP, Hugging Face"
|
||||
},
|
||||
{
|
||||
"@type": "Person",
|
||||
"@id": "https://github.com/waybarrios#person",
|
||||
"name": "Wayner Barrios",
|
||||
"url": "https://github.com/waybarrios",
|
||||
"sameAs": [
|
||||
"https://github.com/waybarrios"
|
||||
]
|
||||
},
|
||||
{
|
||||
"@type": "FAQPage",
|
||||
"@id": "https://skills.waybarrios.com/#faq",
|
||||
"mainEntity": [
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "What is an Agent Skill?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "An Agent Skill is a reusable workflow centered on a SKILL.md file. It provides durable instructions, references, scripts, and decision rules for a specific job."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Does the same skill work in Claude Code, Codex, OpenCode, and Pi?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "The stable workflow IDs are shared, while packaging and activation follow each host's native model."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "What is the difference between code-review and code-reviewer?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "code-review is a comprehensive, merge-oriented review for a pull request or complete change set. code-reviewer is a focused implementation review for small changes and feature-development handoffs."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Can I install only one skill?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "Yes. Run npx @waybarrios/opencode-power-pack install followed by the skill name. The installer also includes required companion skills."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Does the sandbox work with Codex and OpenCode?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "Yes. Codex and OpenCode can explicitly run commands through the shared sandbox CLI on supported Linux and macOS hosts. Automatic routing and whole-agent isolation are not yet implemented."
|
||||
}
|
||||
},
|
||||
{
|
||||
"@type": "Question",
|
||||
"name": "Does the plugin support OpenCode 2.0?",
|
||||
"acceptedAnswer": {
|
||||
"@type": "Answer",
|
||||
"text": "Yes. Git main supports OpenCode 2.0 and OpenCode 1.18.7. The published npm package and tag v0.5.0 predate the V2 plugin port."
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"@type": "ItemList",
|
||||
"@id": "https://skills.waybarrios.com/#profiles",
|
||||
"name": "OpenCode Power Pack installable profiles",
|
||||
"numberOfItems": 8,
|
||||
"itemListElement": [
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 1,
|
||||
"name": "recommended",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Balanced everyday software-engineering workflows."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 2,
|
||||
"name": "review",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Comprehensive, focused, quality, and differential code review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 3,
|
||||
"name": "feature-dev",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Structured feature exploration, architecture, implementation, and review."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 4,
|
||||
"name": "frontend",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Frontend implementation and anti-generic design critique."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 5,
|
||||
"name": "security",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Application-security review, scanning, validation, threat modeling, and reporting."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 6,
|
||||
"name": "huggingface",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Model discovery, inference, training, evaluation, Spaces, and AWS deployment."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 7,
|
||||
"name": "authoring",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Create skills and MCP servers, summarize technical papers, and edit English and Spanish copy on main."
|
||||
},
|
||||
{
|
||||
"@type": "ListItem",
|
||||
"position": 8,
|
||||
"name": "project-memory",
|
||||
"url": "https://skills.waybarrios.com/#profiles",
|
||||
"description": "Durable AGENTS.md and CLAUDE.md project context."
|
||||
}
|
||||
]
|
||||
}
|
||||
]
|
||||
}
|
||||
</script>
|
||||
</head>
|
||||
<body>
|
||||
@@ -181,7 +254,7 @@
|
||||
|
||||
<div class="status-rail" aria-hidden="true">
|
||||
<span>POWER BUS // ONLINE</span>
|
||||
<span>54 MODULES</span>
|
||||
<span>55 MODULES</span>
|
||||
<span>4 RUNTIMES</span>
|
||||
</div>
|
||||
|
||||
@@ -209,13 +282,13 @@
|
||||
<main id="main">
|
||||
<section class="hero" id="top" aria-labelledby="hero-title">
|
||||
<div class="hero-copy">
|
||||
<p class="eyebrow">[ OPEN SOURCE · v0.5.0 · MIT + APACHE-2.0 ]</p>
|
||||
<p class="eyebrow">[ OPEN SOURCE · MAIN · MIT + APACHE-2.0 ]</p>
|
||||
<h1 id="hero-title">
|
||||
One power pack.<br>
|
||||
<span>Every coding agent.</span>
|
||||
</h1>
|
||||
<p class="hero-lede">
|
||||
54 portable Agent Skills for <strong>Claude Code</strong>, <strong>Codex</strong>,
|
||||
55 portable Agent Skills for <strong>Claude Code</strong>, <strong>Codex</strong>,
|
||||
<strong>OpenCode</strong>, and <strong>Pi</strong>. Install the workflows you need for
|
||||
code review, security, feature delivery, frontend craft, MCP authoring, and ML.
|
||||
</p>
|
||||
@@ -299,6 +372,8 @@
|
||||
|
||||
Your goal is to understand what I need, recommend the smallest useful profile or set of individual skills, obtain my explicit selection, install it for the active coding-agent host, preserve all existing project work, and leave a reproducible verification report.
|
||||
|
||||
The npm procedure below uses the published catalog. OpenCode 2 can use those copied skills, but the native V2 plugin and humanizer require Git main. If I need either, show the current Git install guide and wait for my source choice before continuing.
|
||||
|
||||
Safety rules:
|
||||
1. Do not modify package.json, package-lock.json, npm-shrinkwrap.json, pnpm-lock.yaml, yarn.lock, bun.lock, or application dependencies.
|
||||
2. Do not use --force, --global, --all, sudo, destructive Git commands, or manual deletion/replacement of existing skill directories. Do not choose a profile or skill for me, and do not install anything before I explicitly approve the selection.
|
||||
@@ -319,7 +394,7 @@ Phase 2: preflight without changing repository files.
|
||||
|
||||
1. Print the current directory with pwd.
|
||||
2. Run git rev-parse --show-toplevel. Stop if this is not a Git repository. Save the absolute Git root.
|
||||
3. Run node --version. Parse the major version and require Node.js 20 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
3. Run node --version. Parse the full version and require Node.js 20.11.0 or newer. If it is older or unavailable, stop and explain how to satisfy the prerequisite without changing this project.
|
||||
4. Run npm --version and stop if npm or npx is unavailable.
|
||||
5. Capture git status --short before installation. Use it only as a baseline; do not alter any listed file.
|
||||
6. State that the only permitted destination is <git-root>/.agents/skills.
|
||||
@@ -361,8 +436,9 @@ Phase 6: verify and report.
|
||||
3. Run git status --short again and compare it with the saved baseline. Confirm that installation introduced no changes outside .agents/skills and did not change dependency manifests or lockfiles. Do not revert unrelated differences.
|
||||
4. Tell the user to start a new agent session if the skills do not appear automatically.
|
||||
5. Give the host-specific verification command:
|
||||
- Codex: inspect /plugins, then invoke an installed skill such as $code-review.
|
||||
- OpenCode: run opencode debug skill, then invoke /code-review.
|
||||
- Codex: invoke an installed skill such as $code-review in a new session. Use /plugins only for a native plugin installation.
|
||||
- OpenCode 1: run opencode debug skill, then invoke /code-review.
|
||||
- OpenCode 2: check the skill picker, then invoke /code-review.
|
||||
- Pi: run pi list, then request a task that matches an installed workflow.
|
||||
6. Finish with a concise structured report containing: stated need, approved selection and rationale, alternatives not chosen, host, Node/npm versions, Git root, destination, exact commands run, resolved skill count and names, installed entries, preserved/skipped entries, warnings or failures, worktree comparison, verification result, and any user action still required.</pre>
|
||||
<div class="agent-prompt-actions">
|
||||
@@ -387,7 +463,7 @@ Phase 6: verify and report.
|
||||
</div>
|
||||
|
||||
<nav class="hero-routes" aria-label="Explore OpenCode Power Pack">
|
||||
<a href="docs/skills/">Browse all 54 skills <span aria-hidden="true">→</span></a>
|
||||
<a href="docs/skills/">Browse all 55 skills <span aria-hidden="true">→</span></a>
|
||||
<a href="docs/#reviews">Compare review workflows <span aria-hidden="true">→</span></a>
|
||||
<a href="docs/#how-skills-work">Learn how Agent Skills work <span aria-hidden="true">→</span></a>
|
||||
<a href="docs/#npm">Installation examples <span aria-hidden="true">→</span></a>
|
||||
@@ -406,12 +482,22 @@ Phase 6: verify and report.
|
||||
</section>
|
||||
|
||||
<section class="proof-strip" aria-label="Project facts">
|
||||
<div><strong>54</strong><span>curated skills</span></div>
|
||||
<div><strong>55</strong><span>curated skills</span></div>
|
||||
<div><strong>4</strong><span>agent hosts</span></div>
|
||||
<div><strong>400</strong><span>automated tests</span></div>
|
||||
<div><strong>418</strong><span>automated tests</span></div>
|
||||
<div><strong>4</strong><span>sandbox profiles</span></div>
|
||||
</section>
|
||||
|
||||
<section class="section update-section" id="whats-new" aria-labelledby="updates-title">
|
||||
<div class="section-heading">
|
||||
<p class="kicker">New on main</p>
|
||||
<h2 id="updates-title">OpenCode 2.0 and humanizer.</h2>
|
||||
<p>OpenCode 1.18.7, 2.0.14, and 2.0.32 load all 55 skills and three specialists. V2 permission checks keep the bundled specialists read-only.</p>
|
||||
<p><a href="docs/skills/#humanizer">Humanizer</a> edits English and Spanish copy without changing the facts. The authoring profile now has four skills.</p>
|
||||
<p class="release-note">The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use Git main for OpenCode 2 and humanizer. <a href="docs/#opencode-2">See the install commands.</a></p>
|
||||
</div>
|
||||
</section>
|
||||
|
||||
<section class="section install-section" id="profiles" aria-labelledby="install-title">
|
||||
<div class="section-index">01 / PROFILES</div>
|
||||
<div class="section-heading">
|
||||
@@ -479,12 +565,12 @@ Phase 6: verify and report.
|
||||
</article>
|
||||
|
||||
<article class="profile-card">
|
||||
<div><span>AUTHORING</span><b>3 skills</b></div>
|
||||
<div><span>AUTHORING</span><b>4 skills on main</b></div>
|
||||
<h3>Reusable agent tools</h3>
|
||||
<p><strong>For:</strong> builders creating new Agent Skills, MCP servers, or engineering summaries from technical research.</p>
|
||||
<small>Includes skill-creator, mcp-builder, and paper-summarizer.</small>
|
||||
<code>npx @waybarrios/opencode-power-pack install --profile authoring</code>
|
||||
<button type="button" class="small-copy" data-copy="npx @waybarrios/opencode-power-pack install --profile authoring">Copy command</button>
|
||||
<p><strong>For:</strong> builders creating new Agent Skills, MCP servers, technical-paper summaries, and clear English or Spanish copy.</p>
|
||||
<small>Includes humanizer, skill-creator, mcp-builder, and paper-summarizer on main.</small>
|
||||
<code>npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install --profile authoring</code>
|
||||
<button type="button" class="small-copy" data-copy="npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install --profile authoring">Copy command</button>
|
||||
</article>
|
||||
|
||||
<article class="profile-card">
|
||||
@@ -524,9 +610,9 @@ Phase 6: verify and report.
|
||||
</article>
|
||||
<article class="platform-card platform-opencode">
|
||||
<span class="platform-number">C</span>
|
||||
<h3>OpenCode</h3>
|
||||
<p>Native plugin loading plus least-privilege feature-development specialists.</p>
|
||||
<code>opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"</code>
|
||||
<h3>OpenCode 1 + 2</h3>
|
||||
<p>OpenCode 2.0 support on main, with three read-only specialists. OpenCode 1.18.7 still works.</p>
|
||||
<code>opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"</code>
|
||||
</article>
|
||||
<article class="platform-card platform-pi">
|
||||
<span class="platform-number">D</span>
|
||||
@@ -542,12 +628,12 @@ Phase 6: verify and report.
|
||||
<div class="section-heading catalog-heading">
|
||||
<div>
|
||||
<p class="kicker">Searchable skill inventory</p>
|
||||
<h2 id="skills-title">54 workflows. Three operational bands.</h2>
|
||||
<h2 id="skills-title">55 workflows. Three operational bands.</h2>
|
||||
</div>
|
||||
<div class="skill-search">
|
||||
<label for="skill-filter">Filter skills</label>
|
||||
<input id="skill-filter" type="search" placeholder="Try: review, semgrep, huggingface">
|
||||
<p id="filter-status" aria-live="polite">Showing all 54 skills</p>
|
||||
<p id="filter-status" aria-live="polite">Showing all 55 skills</p>
|
||||
<a class="catalog-reference-link" href="docs/skills/">Descriptions + install commands →</a>
|
||||
</div>
|
||||
</div>
|
||||
@@ -557,7 +643,7 @@ Phase 6: verify and report.
|
||||
<div class="group-header">
|
||||
<span>CORE</span>
|
||||
<h3 id="core-title">Build, review, and author</h3>
|
||||
<b>14</b>
|
||||
<b>15</b>
|
||||
</div>
|
||||
<ul class="skill-list">
|
||||
<li data-skill="agents-md-improver project memory rules"><a href="docs/skills/#agents-md-improver">agents-md-improver</a></li>
|
||||
@@ -571,6 +657,7 @@ Phase 6: verify and report.
|
||||
<li data-skill="design-patterns architecture"><a href="docs/skills/#design-patterns">design-patterns</a></li>
|
||||
<li data-skill="feature-dev implementation workflow"><a href="docs/skills/#feature-dev">feature-dev</a></li>
|
||||
<li data-skill="frontend-design web interface accessibility"><a href="docs/skills/#frontend-design">frontend-design</a></li>
|
||||
<li data-skill="humanizer writing copy editing english spanish"><a href="docs/skills/#humanizer">humanizer</a></li>
|
||||
<li data-skill="mcp-builder model context protocol"><a href="docs/skills/#mcp-builder">mcp-builder</a></li>
|
||||
<li data-skill="paper-summarizer research"><a href="docs/skills/#paper-summarizer">paper-summarizer</a></li>
|
||||
<li data-skill="skill-creator agent skills authoring"><a href="docs/skills/#skill-creator">skill-creator</a></li>
|
||||
@@ -646,15 +733,15 @@ Phase 6: verify and report.
|
||||
<p class="kicker">Curated means verified</p>
|
||||
<h2 id="quality-title">Portable workflows with an evidence trail.</h2>
|
||||
<p>
|
||||
This is not a bulk mirror. Skills are reviewed, adapted for cross-host behavior, and
|
||||
shipped with provenance, notices, licenses, host validation, and regression coverage.
|
||||
Each workflow has source records and checks for the hosts it supports. The repository is
|
||||
shipped with provenance, notices, licenses, host checks, and regression tests.
|
||||
</p>
|
||||
</div>
|
||||
<ol class="quality-list">
|
||||
<li><b>01</b><span><strong>54 / 54</strong> official Agent Skills validations</span></li>
|
||||
<li><b>02</b><span><strong>400</strong> automated behavioral, packaging, and sandbox tests</span></li>
|
||||
<li><b>01</b><span><strong>55 / 55</strong> official Agent Skills validations</span></li>
|
||||
<li><b>02</b><span><strong>418</strong> automated behavioral, packaging, and sandbox tests</span></li>
|
||||
<li><b>03</b><span><strong>Strict</strong> Claude plugin and marketplace validation</span></li>
|
||||
<li><b>04</b><span><strong>Minimum + latest</strong> OpenCode smoke coverage</span></li>
|
||||
<li><b>04</b><span><strong>OpenCode 1 + 2</strong> OpenCode smoke coverage</span></li>
|
||||
<li><b>05</b><span><strong>Immutable</strong> upstream provenance and license records</span></li>
|
||||
</ol>
|
||||
</div>
|
||||
@@ -690,6 +777,10 @@ Phase 6: verify and report.
|
||||
small changes and feature-development handoffs.
|
||||
</p>
|
||||
</details>
|
||||
<details>
|
||||
<summary>Does the plugin support OpenCode 2.0?</summary>
|
||||
<p>Yes. Install from Git main with <code>opencode plugin add</code>. The same plugin supports OpenCode 1.18.7 through <code>plugin --global</code>. npm 0.5.0 predates the V2 port. <a href="docs/#opencode-2">Read the compatibility guide.</a></p>
|
||||
</details>
|
||||
<details>
|
||||
<summary>Can I install only one skill?</summary>
|
||||
<p>
|
||||
@@ -737,9 +828,9 @@ Phase 6: verify and report.
|
||||
<a href="llms.txt">AI summary</a>
|
||||
<a href="https://github.com/waybarrios/opencode-power-pack/blob/main/LICENSE">License</a>
|
||||
</nav>
|
||||
<span>v0.5.0 // 2026</span>
|
||||
<span>MAIN // 2026</span>
|
||||
</footer>
|
||||
|
||||
<script src="app.js?v=0.5.0-sandbox-1" defer></script>
|
||||
<script src="app.js?v=2026-10-19-opencode-2" defer></script>
|
||||
</body>
|
||||
</html>
|
||||
|
||||
@@ -1,6 +1,14 @@
|
||||
# OpenCode Power Pack
|
||||
|
||||
OpenCode Power Pack is an open-source collection of 54 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi. The current release is v0.5.0.
|
||||
OpenCode Power Pack is an open-source collection of 55 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi. This page follows main. The latest published release is v0.5.0.
|
||||
|
||||
## OpenCode 2.0 and humanizer
|
||||
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
OpenCode 2 uses `opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`. OpenCode 1 uses `opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`. Both load 55 skills and three specialists from main.
|
||||
|
||||
Humanizer edits English and Spanish copy without changing facts, names, numbers, URLs, or code. It joins mcp-builder, paper-summarizer, and skill-creator in the four-skill authoring profile.
|
||||
|
||||
## Start here
|
||||
|
||||
@@ -12,7 +20,7 @@ OpenCode Power Pack is an open-source collection of 54 portable Agent Skills for
|
||||
|
||||
## Install
|
||||
|
||||
The selective npm installer requires Node.js 20 or newer.
|
||||
The selective npm installer requires Node.js 20.11.0 or newer.
|
||||
|
||||
```sh
|
||||
# See every skill and profile
|
||||
@@ -40,7 +48,7 @@ To delegate installation to a coding agent, use the safe dry-run-first prompt an
|
||||
| `frontend` | Frontend developers and designers | Interface implementation and anti-generic design critique |
|
||||
| `security` | AppSec engineers and security-conscious maintainers | Review, threat modeling, scanning, validation, and reporting |
|
||||
| `huggingface` | ML engineers and researchers | Discovery, inference, training, evaluation, Spaces, and AWS deployment |
|
||||
| `authoring` | Builders of reusable agent tools | Agent Skills, MCP servers, and paper summarization |
|
||||
| `authoring` | Builders of reusable agent tools | Agent Skills, MCP servers, paper summaries, and copy editing on main |
|
||||
| `project-memory` | Teams maintaining durable agent context | AGENTS.md and CLAUDE.md improvement and revision |
|
||||
|
||||
Install a profile with:
|
||||
@@ -53,7 +61,7 @@ npx @waybarrios/opencode-power-pack install --profile <profile-name>
|
||||
|
||||
- Claude Code uses namespaced plugin skills such as `/opencode-power-pack:code-review`.
|
||||
- Codex invokes installed workflows with names such as `$code-review`.
|
||||
- OpenCode invokes installed workflows with names such as `/code-review`.
|
||||
- OpenCode 1 and 2 invoke installed workflows with names such as `/code-review`. Install the current Git plugin for OpenCode 2.
|
||||
- Pi exposes the package's declared skills directory.
|
||||
|
||||
## Review workflows
|
||||
@@ -80,10 +88,10 @@ Codex, OpenCode, Claude Code, and Pi can all invoke the manual runner. Automatic
|
||||
|
||||
## Quality signals
|
||||
|
||||
- 54 of 54 official Agent Skills validations pass.
|
||||
- 400 automated behavioral, packaging, and sandbox tests are defined at v0.5.0.
|
||||
- 55 of 55 official Agent Skills validations pass.
|
||||
- The main branch defines 418 tests. The latest local run passed 417 and skipped the optional native sandbox integration test.
|
||||
- Claude plugin and marketplace metadata receive strict validation.
|
||||
- OpenCode has minimum-version and latest-version smoke coverage.
|
||||
- Smoke tests cover OpenCode 1.18.7, 2.0.14, and 2.0.32, including effective V2 permission checks.
|
||||
- Third-party workflows include immutable provenance, notices, and license records.
|
||||
|
||||
OpenCode Power Pack is MIT AND Apache-2.0 licensed, subject to the per-skill provenance and notices in the repository.
|
||||
|
||||
+23
-14
@@ -1,8 +1,8 @@
|
||||
# OpenCode Power Pack: Expanded Documentation
|
||||
|
||||
> Canonical, machine-readable project facts for OpenCode Power Pack v0.5.0. Last updated 2026-08-17.
|
||||
> Canonical, machine-readable project facts for the current main branch of OpenCode Power Pack. Last updated 2026-10-19.
|
||||
|
||||
OpenCode Power Pack is an open-source collection of 54 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi. It packages repeatable workflows for software development, code review, application security, frontend design, MCP authoring, project memory, and Hugging Face machine-learning work.
|
||||
OpenCode Power Pack is an open-source collection of 55 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi. It packages repeatable workflows for software development, code review, application security, frontend design, MCP authoring, project memory, and Hugging Face machine-learning work.
|
||||
|
||||
Canonical website: https://skills.waybarrios.com/
|
||||
Documentation: https://skills.waybarrios.com/docs/
|
||||
@@ -13,13 +13,21 @@ Release: https://github.com/waybarrios/opencode-power-pack/releases/tag/v0.5.0
|
||||
Author: Wayner Barrios
|
||||
License: MIT AND Apache-2.0, with per-skill provenance and notices
|
||||
|
||||
## OpenCode 2.0 and humanizer
|
||||
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
OpenCode 2 install: `opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`. OpenCode 1 install: `opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`.
|
||||
|
||||
Both hosts load all 55 skills and three specialists from main. Install humanizer selectively with `npx --package=git+https://github.com/waybarrios/opencode-power-pack.git opencode-power-pack install humanizer`. Invoke it with `$humanizer` in Codex, `/humanizer` in OpenCode, or `/opencode-power-pack:humanizer` in Claude Code. It edits English and Spanish copy while keeping facts, names, numbers, URLs, and code intact.
|
||||
|
||||
## What an Agent Skill is
|
||||
|
||||
An Agent Skill is a reusable workflow centered on a SKILL.md file. It can include durable instructions, decision rules, references, scripts, and examples for a specific task. The workflow ID stays stable across supported coding agents, while packaging and invocation follow each host's conventions.
|
||||
|
||||
## Selective npm installation
|
||||
|
||||
Requirement: Node.js 20 or newer.
|
||||
Requirement: Node.js 20.11.0 or newer.
|
||||
|
||||
List skills and profiles:
|
||||
|
||||
@@ -45,7 +53,7 @@ Install the recommended profile:
|
||||
|
||||
npx @waybarrios/opencode-power-pack install --profile recommended
|
||||
|
||||
Install all 54 skills:
|
||||
Install every skill in the published npm catalog:
|
||||
|
||||
npx @waybarrios/opencode-power-pack install --all
|
||||
|
||||
@@ -100,7 +108,7 @@ Compatibility specification: https://github.com/waybarrios/opencode-power-pack/b
|
||||
- frontend: frontend-design and ai-slop.
|
||||
- security: 15 security review, threat-modeling, static-analysis, validation, and reporting workflows.
|
||||
- huggingface: 25 discovery, inference, training, evaluation, Spaces, and AWS deployment workflows.
|
||||
- authoring: mcp-builder, paper-summarizer, and skill-creator.
|
||||
- authoring on main: humanizer, mcp-builder, paper-summarizer, and skill-creator. npm 0.5.0 contains the three earlier authoring skills.
|
||||
- project-memory: agents-md-improver and agents-md-revise.
|
||||
|
||||
Install any profile with:
|
||||
@@ -117,9 +125,9 @@ Install any profile with:
|
||||
|
||||
## Complete skill inventory
|
||||
|
||||
Every skill can be installed with `npx @waybarrios/opencode-power-pack install <skill-id>`. Purpose, dependencies, profiles, host invocation, and source links are available at `https://skills.waybarrios.com/docs/skills/#<skill-id>`.
|
||||
Released skills can be installed with `npx @waybarrios/opencode-power-pack install <skill-id>`. Purpose, dependencies, profiles, host invocation, and source links, including the Git-only humanizer install, are available at `https://skills.waybarrios.com/docs/skills/#<skill-id>`.
|
||||
|
||||
### Core and authoring skills (14)
|
||||
### Core and authoring skills (15)
|
||||
|
||||
1. agents-md-improver: audits and improves AGENTS.md, CLAUDE.md, and related project-rules files.
|
||||
2. agents-md-revise: captures session learnings into durable project rules.
|
||||
@@ -132,9 +140,10 @@ Every skill can be installed with `npx @waybarrios/opencode-power-pack install <
|
||||
9. design-patterns: evaluates pattern trade-offs for Rust, TypeScript, React, Django, and Python.
|
||||
10. feature-dev: coordinates a structured end-to-end feature implementation workflow.
|
||||
11. frontend-design: creates distinctive, production-grade frontend interfaces.
|
||||
12. mcp-builder: designs and builds Model Context Protocol servers.
|
||||
13. paper-summarizer: analyzes academic and technical papers for engineering decisions.
|
||||
14. skill-creator: creates or improves reusable Agent Skills and their trigger descriptions.
|
||||
12. humanizer: rewrites English and Spanish copy while preserving facts, names, numbers, URLs, and code.
|
||||
13. mcp-builder: designs and builds Model Context Protocol servers.
|
||||
14. paper-summarizer: analyzes academic and technical papers for engineering decisions.
|
||||
15. skill-creator: creates or improves reusable Agent Skills and their trigger descriptions.
|
||||
|
||||
### Security skills (15)
|
||||
|
||||
@@ -184,10 +193,10 @@ Every skill can be installed with `npx @waybarrios/opencode-power-pack install <
|
||||
|
||||
## Quality and provenance
|
||||
|
||||
- 54 of 54 skills pass official Agent Skills validation.
|
||||
- The repository defines 400 automated behavioral, packaging, and sandbox tests at v0.5.0.
|
||||
- 55 of 55 skills pass official Agent Skills validation.
|
||||
- The main branch defines 418 tests. The latest local run passed 417 and skipped the optional native sandbox integration.
|
||||
- Claude plugin and marketplace metadata receive strict validation.
|
||||
- OpenCode receives minimum-version and latest-version smoke coverage.
|
||||
- OpenCode smoke tests cover 1.18.7, 2.0.14, and 2.0.32, including effective read-only permissions on V2.
|
||||
- Third-party workflows have immutable upstream references, notices, and license records.
|
||||
|
||||
Evidence and exact provenance live in the repository:
|
||||
@@ -199,7 +208,7 @@ Evidence and exact provenance live in the repository:
|
||||
## Citation and factual boundaries
|
||||
|
||||
Preferred project name: OpenCode Power Pack.
|
||||
Current documented version: 0.5.0.
|
||||
Current documentation: main. Latest published release: 0.5.0.
|
||||
Canonical URL: https://skills.waybarrios.com/
|
||||
Repository: https://github.com/waybarrios/opencode-power-pack
|
||||
Package: @waybarrios/opencode-power-pack
|
||||
|
||||
@@ -1,8 +1,8 @@
|
||||
# OpenCode Power Pack
|
||||
|
||||
> OpenCode Power Pack v0.5.0 is an open-source collection of 54 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi, with an opt-in native sandbox runner for contained commands.
|
||||
> OpenCode Power Pack on main is an open-source collection of 55 portable Agent Skills for Claude Code, Codex, OpenCode, and Pi, with an opt-in native sandbox runner for contained commands.
|
||||
|
||||
The project provides reusable workflows for code review, security auditing, feature development, frontend design, MCP authoring, project memory, and Hugging Face machine-learning work. Its npm package is `@waybarrios/opencode-power-pack`. The selective installer requires Node.js 20 or newer and provides eight curated profiles.
|
||||
The project provides reusable workflows for code review, security auditing, feature development, frontend design, MCP authoring, project memory, and Hugging Face machine-learning work. Its npm package is `@waybarrios/opencode-power-pack`. The selective installer requires Node.js 20.11.0 or newer and provides eight curated profiles.
|
||||
|
||||
Install one skill with `npx @waybarrios/opencode-power-pack install <skill-id>`. Install the recommended profile with `npx @waybarrios/opencode-power-pack install --profile recommended`. List every skill and profile with `npx @waybarrios/opencode-power-pack list`.
|
||||
|
||||
@@ -10,21 +10,29 @@ The `code-review` workflow performs a comprehensive review of a pull request or
|
||||
|
||||
The manual sandbox runner works from Codex, OpenCode, Claude Code, and Pi on supported macOS and Linux hosts. It contains the explicitly selected command and its descendants. Automatic host routing and whole-agent isolation are not yet implemented.
|
||||
|
||||
When citing the project, use “OpenCode Power Pack,” version 0.5.0, author Wayner Barrios, canonical URL `https://skills.waybarrios.com/`, and repository `https://github.com/waybarrios/opencode-power-pack`.
|
||||
When citing the project, use “OpenCode Power Pack,” the main branch or a specific release tag, author Wayner Barrios, canonical URL `https://skills.waybarrios.com/`, and repository `https://github.com/waybarrios/opencode-power-pack`.
|
||||
|
||||
## OpenCode 2.0 and humanizer
|
||||
|
||||
The current main branch includes 55 skills, OpenCode 2.0 support, and humanizer. The published npm package and tag v0.5.0 still contain 54 skills and the V1-only plugin. Use the Git installation for the full OpenCode 2 plugin and humanizer.
|
||||
|
||||
OpenCode 2 install: `opencode plugin add "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`. OpenCode 1 install: `opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"`.
|
||||
|
||||
Both hosts load all 55 skills and three specialists from main. Invoke humanizer with `$humanizer` in Codex, `/humanizer` in OpenCode, or `/opencode-power-pack:humanizer` in Claude Code. It edits English and Spanish copy while keeping facts, names, numbers, URLs, and code intact.
|
||||
|
||||
## Documentation
|
||||
|
||||
- [Project overview](https://skills.waybarrios.com/index.md): Product purpose, supported hosts, install paths, profiles, and quality signals.
|
||||
- [Installation, sandbox, and operations](https://skills.waybarrios.com/docs/index.md): npm commands, host setup, contained command execution, profile selection, updates, removal, and troubleshooting.
|
||||
- [Agent installation guide](https://skills.waybarrios.com/agent-install.md): Safe dry-run-first instructions and a copyable prompt for an agent to perform the installation.
|
||||
- [Complete 54-skill reference](https://skills.waybarrios.com/docs/skills/index.md): Purpose, audience, profile membership, dependencies, install command, and source link for every skill.
|
||||
- [Complete 55-skill reference](https://skills.waybarrios.com/docs/skills/index.md): Purpose, audience, profile membership, dependencies, install command, and source link for every skill.
|
||||
- [Expanded project reference](https://skills.waybarrios.com/llms-full.txt): Single-file project facts, commands, profiles, inventory, provenance, and citation guidance.
|
||||
|
||||
## Source and distribution
|
||||
|
||||
- [GitHub repository](https://github.com/waybarrios/opencode-power-pack#readme): Canonical source code and human documentation.
|
||||
- [npm package](https://www.npmjs.com/package/@waybarrios/opencode-power-pack): Public selective installer package.
|
||||
- [Release v0.5.0](https://github.com/waybarrios/opencode-power-pack/releases/tag/v0.5.0): Current documented release.
|
||||
- [Release v0.5.0](https://github.com/waybarrios/opencode-power-pack/releases/tag/v0.5.0): Latest published release, before OpenCode 2 support and humanizer.
|
||||
- [Upstream provenance](https://github.com/waybarrios/opencode-power-pack/blob/main/UPSTREAMS.json): Immutable source references for adapted workflows.
|
||||
|
||||
## Optional
|
||||
|
||||
+1
-1
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "OpenCode Power Pack",
|
||||
"short_name": "Power Pack",
|
||||
"description": "Fifty-four portable Agent Skills for Claude Code, Codex, OpenCode, and Pi.",
|
||||
"description": "Fifty-five portable Agent Skills for Claude Code, Codex, OpenCode, and Pi.",
|
||||
"start_url": "/",
|
||||
"scope": "/",
|
||||
"display": "standalone",
|
||||
|
||||
+4
-4
@@ -5,23 +5,23 @@
|
||||
>
|
||||
<url>
|
||||
<loc>https://skills.waybarrios.com/</loc>
|
||||
<lastmod>2026-08-17</lastmod>
|
||||
<lastmod>2026-10-19</lastmod>
|
||||
<changefreq>weekly</changefreq>
|
||||
<priority>1.0</priority>
|
||||
<image:image>
|
||||
<image:loc>https://skills.waybarrios.com/assets/social-card.png</image:loc>
|
||||
<image:title>OpenCode Power Pack: 54 skills for four coding agents</image:title>
|
||||
<image:title>OpenCode Power Pack: 55 skills for four coding agents</image:title>
|
||||
</image:image>
|
||||
</url>
|
||||
<url>
|
||||
<loc>https://skills.waybarrios.com/docs/</loc>
|
||||
<lastmod>2026-08-17</lastmod>
|
||||
<lastmod>2026-10-19</lastmod>
|
||||
<changefreq>weekly</changefreq>
|
||||
<priority>0.9</priority>
|
||||
</url>
|
||||
<url>
|
||||
<loc>https://skills.waybarrios.com/docs/skills/</loc>
|
||||
<lastmod>2026-08-17</lastmod>
|
||||
<lastmod>2026-10-19</lastmod>
|
||||
<changefreq>weekly</changefreq>
|
||||
<priority>0.9</priority>
|
||||
</url>
|
||||
|
||||
Reference in New Issue
Block a user