The add-field menu listed bare keys such as `context` and `shell`, so
picking one meant already knowing what it does. Each option now shows the
same hint the field shows once it is added, so no new strings are needed.
Context menu items take an optional description; menus that do not pass
one render as before.
Claude Code takes a whole server entry from one scope, so the lone switch
used for OpenCode, Kilo Code and Pi would replace the server in .mcp.json
instead of turning it off. In project mode a disabled entry for claude now
adds the name to that project's disabledMcpServers list in ~/.claude.json,
the list /mcp edits. It keys apart in the ledger from a user-scope server
of the same name that a global config may own in the same file, and a name
the person turned off themselves is never claimed or removed.
A local scope server of the same name silently wins over what Skillshare
writes, so project-mode plans now report it with the command to remove it.
Checks against each Agent's documentation also showed gaps:
- Detection missed Agents whose MCP file does not exist yet, such as a
fresh Claude Code install, and listed every Agent in project mode.
It now looks at the Agent's own folder and at project files.
- Cline moved its settings to ~/.cline/data and honors CLINE_DIR,
CLINE_DATA_DIR and CLINE_MCP_SETTINGS_PATH; the old VS Code path is
kept only while it is the one in use.
- Kilo Code ignores a project config that holds an environment
reference, and Claude Code skips reserved names and reads its own
credentials as empty for remote servers. These are refused up front,
and the dashboard preview refuses them too.
- An entry owned by a config that no longer exists could never be
released; an explicit resolution may now take it over.
- --disabled was accepted by commands that ignored it, and a relative
XDG_CONFIG_HOME or Cline override was used as given.
Refs #286
The import dialog listed Agents by name only, while every other MCP
surface shows the Agent's logo, so the picker was harder to scan. Select
options can now carry an icon. Claude Desktop gets the Claude mark instead
of the generic fallback.
The Agent toggles in both MCP dialogs now update from the previous state,
so two quick clicks cannot drop one of the changes.
A server defined in an Agent's global config could not be switched off for
a single project without editing that Agent's file by hand. A source entry
with only 'disabled: true' now writes the Agent's own off switch into its
project file. Only Agents that merge project entries over global ones field
by field can honour a lone switch, so this is limited to OpenCode, Kilo Code
and Pi with pi-mcp-adapter; other targets and global mode are refused with
an explanation rather than writing an entry that would break the server.
Refs #286
Factory fell back to a letter and LM Studio to a grey glyph, so both were
hard to spot among the Agent toggles. Factory uses its official mark as a
mono mask; LM Studio uses its purple gradient badge.
Kilo Code is an OpenCode fork, so its MCP settings use the same "mcp"
section, local/remote types and {env:VAR} references. Reuse the OpenCode
rendering and import paths instead of adding a second format.
Kilo merges every config file it finds (kilo.jsonc and kilo.json, in the
project root and in .kilo/), so an existing file is reused, a new one is
created as kilo.jsonc, and more than one existing file is refused: entries
in one would mask the other.
The kilocode skills target moves from .kilocode/skills to .kilo/skills,
which is the only location Kilo's documentation lists now. Users who
already synced kilocode need to run sync again.
Also give kilocode its brand-colored icon so it is not mistaken for Kiro,
and add the missing icon for omp.
Refs #287
* feat(mcp): sync portable server configuration across six clients
Add inline and external MCP sources with guided CLI and dashboard setup, native imports, conflict previews, ownership tracking, backups, and restore. Support Claude Code, Codex, Cursor, VS Code, OpenCode, and Grok CLI in global and project scopes.
Format YAML on save, document MCP fields in the config assistant, and keep its layout stable for scalar list items. Include schemas, localized UI copy, documentation, and regression and end-to-end coverage.
* fix(config): keep loading config when the MCP section is invalid
A mistake in the mcp section (an unknown key, a targets value that is
not a list, an invalid server field) made config.Load, LoadProject,
Save, SaveIn and ValidateConfig fail. Every command that reads config,
including skills sync, install, list and the dashboard, stopped working
because of settings it never uses.
MCPConfig now keeps a section that does not parse exactly as written and
records the error instead of returning it, so commands unrelated to MCP
load and save the config without touching that section. MCP commands
still validate their source on every run, and the dashboard config
editor calls the new ValidateMCP, so saving an invalid MCP section there
is still rejected.
Also correct the ProjectSources comment: sources.mcp resolves from the
directory containing config.yaml, not from the project root.
* fix(mcp): protect Agent files during concurrent edits and recovery
A release review of the MCP sync engine found ways it could lock users
out, report false conflicts, or change entries it does not own.
Recovery
- A failed write left pending.json behind, and every later MCP command
refused to run. Failed writes now remove their journal, and a journal
whose Agent file changed again is discarded, because ownership hashes
still flag entries that no longer match.
- Preview no longer writes. It reads the ownership that recovery will
record, byte for byte, so a preview stays valid across the recovery
that the next write performs under the lock.
Concurrent Agent writes
- Claude Code rewrites ~/.claude.json on its own, so hashing whole files
expired previews and failed applies. Revisions and pre-write checks
now cover only the MCP entries, and apply rereads each file and applies
the previewed entry edits to its current content, keeping settings the
Agent wrote after the preview.
Comparison and ownership
- Only fields Skillshare renders are hashed, after dropping defaults
Agents add or omit (type stdio/http, empty env, args or headers,
enabled: true) and header name case, so equivalent entries are not
reported as drift. enabled and disabled are compared, so turning off a
managed server is a conflict instead of being overwritten.
- An unmanaged entry that already matches the source is reported as
unchanged but no longer claimed, so removing the server never deletes
an entry the user wrote. Import adopts it explicitly, and without
--replace refuses to rewrite a differing entry, such as a literal token
that import turned into an environment reference.
- Sync keeps Agent-only fields such as timeouts in existing entries.
Import
- Agent-only fields become warnings instead of blocking the import.
- streamable-http is accepted, and enabled/disabled flags are checked
for every client.
- Values carrying a URL password, such as database DSNs, become
environment references, and credential-like arguments are flagged.
Writes and state
- Save only renders the definition first, so a server that could never
sync, such as one without targets, is rejected before it is saved.
- A symlinked config.yaml is written through to its target.
- TOML entries are updated in place, keep CRLF line endings, and repeated
add/remove cycles no longer grow blank lines.
- JSON edits no longer HTML-escape characters such as &.
- GROK_HOME is honored.
- Backups keep the newest 20 per Agent file, pruned by file name, and
damaged backup files are skipped instead of failing the list.
- GET /api/mcp no longer fails when an Agent that no server uses has an
unresolvable path.
* feat(mcp): detect the client format of pasted JSON
Import without --from, the add wizard and the dashboard import API
assumed Claude's mcpServers shape unless a client was named, so VS Code
"servers" and OpenCode "mcp" snippets copied from server docs failed to
import. The client is now chosen by the top-level key, and a bare server
object still reads as Claude's single-entry shape.
TOML stays explicit because Codex and Grok share the format; a .toml
file without --from is still read as Codex.
* refactor(mcp): print CLI results through internal/ui
MCP commands wrote plain fmt output and duplicated statusFromErr, unlike
every other command. Plans, backup IDs and save/apply messages now use
the shared ui helpers, so they get the standard status prefixes and
colors; each change line shows the server name, action and target. The
operation log uses statusFromErr. --json output is unchanged.
* fix(server): serve MCP settings only to localhost or IP hosts
MCP settings include stdio commands that Agents execute, so a web page
must not be able to read or change them. The existing Origin check does
not stop DNS rebinding: once an attacker's domain resolves to 127.0.0.1,
the browser sends that domain as both Origin and Host from a loopback
address, and the dashboard also binds 0.0.0.0 in Docker.
Rebinding always needs a domain name, so every MCP route now also
requires a Host that is localhost or an IP literal, whatever the bind
address. Other dashboard routes are unaffected. Opening the dashboard
through a domain name, including a reverse proxy, therefore returns 403
for MCP settings; the message says to use localhost or an IP address,
because rewriting Host at the proxy would still fail the Origin check
and would reopen the hole.
Remove the unused POST /api/mcp/sync endpoint. Configuring with sync
covers it, and the guard wraps every remaining MCP route.
* fix(sync): make sync --all report MCP results reliably
- "sync -g mcp" and "sync --dry-run mcp" ran skills sync, because mcp
was only recognized as the first argument. It now matches in any
position.
- sync --all --json printed plain-text errors when the MCP preflight
failed, and only an error on conflicts. Preflight errors now use the
JSON error shape, and conflicts also return the credential-free plan so
scripts can see which entries conflict.
- When resource sync failed, MCP was skipped without notice. Text output
now warns that MCP settings were not applied; JSON output omits the mcp
field, which --all otherwise always includes.
- MCP was applied with the preflight revision, which failed whenever the
resource sync changed config or state in between. It is now planned
again under the MCP lock, and conflicts found then still stop the
write.
* fix(sync): recognize agents and extras after flags
sync only checked the first argument for a resource kind, so
"sync --dry-run agents" and "sync -g extras" silently synchronized
skills instead. Sync flags take no values, so the kind now matches in
any position.
The shared parseKindArg keeps its first-argument rule: uninstall, update
and similar commands accept resource names, and a name such as "agents"
must not be read as a kind.
* feat(ui): redesign the MCP dashboard around a server-by-agent matrix
The MCP page listed servers and planned changes as flat text, which made
it hard to see where each server is written and what needs attention.
- A matrix shows each server's status per Agent, summarizes pending
changes and conflicts, and can hide synced entries. Agents whose MCP
file or config directory exists are marked as detected; GET /api/mcp
now returns the detected clients.
- Setup is a two-step dialog: add a remote URL, paste JSON or TOML, or
import from an Agent, then preview which Agent files change. Importing
an entry the Agent already has adopts it, Replace overrides only that
entry, and a target selection equal to the inherited mcp.targets is
saved without pinning targets on the server.
- Conflicts explain their cause and offer Import from <agent> or
Replace. Removal lists the affected Agent files and offers removing
from the source only or removing and syncing. Backups are grouped by
day, with a preview before restoring a file.
- Agent icons appear on the MCP and Targets pages, and the universal
target is marked as a path shared across agents. Brand icons come from
@lobehub/icons-static-svg, with bundled SVGs for Warp and Zed.
- Segmented controls wrap instead of overflowing narrow containers.
- New copy is translated for every supported locale.
* fix(ui): let Sync all apply MCP after syncing resources
Sync all previewed MCP, synced skills, agents and extras, then applied
MCP with the revision from the first preview. When the resource sync
changed config or state, that revision was stale and MCP failed with
"preview again" after everything else had already been synced.
MCP is now rechecked before and after the resource sync by comparing the
reviewed changes (target, name and action) instead of the revision, and
applied with the fresh revision. A different set of changes still stops
before any MCP file is written.
* fix(ui): keep flow-style YAML collections when saving config
Formatting on save forced every collection into block style, so a
hand-written "targets: [claude, codex]" was expanded into a list each
time the config was saved. Each collection now keeps its flow or block
style, indentation is still normalized and comments are preserved, and
flow collections are not padded with spaces.
* docs(mcp): document ownership, normalization and dashboard host rules
- Explain that matching entries stay unmanaged until imported, so
removing a server leaves them in place, and that import adopts an
identical entry while --replace rewrites a differing one.
- Note that Agent defaults and header case are not treated as changes,
that disabling a managed server is a conflict, and that previews stay
valid while an Agent rewrites unrelated settings.
- Describe Agent-specific fields, streamable-http, GROK_HOME, backup
retention, recovery by the next MCP write, and the VS Code limitation
on environment references in headers.
- Document that MCP settings require opening the dashboard by localhost
or an IP address, with SSH port forwarding for remote machines.
- Update the built-in skill for all six clients and the adoption rule,
and the MCP runbook for import refusal, Agent-specific fields and
pulled changes.
* docs(skill): streamline guidance and refresh resource workflows
Keep the built-in skill entrypoint focused and route agent and MCP details to dedicated references. Correct scope, audit, automation, and recovery guidance against the current CLI.
* feat(ui): redesign the dashboard around two styles and a flatter layout
The dashboard had grown page by page, so the same idea looked different
depending on where you met it and pages held controls with nothing to act on.
This rebuilds it on one token set that drives both the Clean and Playful styles
in light and dark, and reshapes the pages around what a person actually does.
Notable changes:
- MCP: replace the server-by-agent matrix with one row per server. The agents a
server writes to are chips inside the row, so the list grows downwards as more
CLIs gain MCP support instead of scrolling sideways.
- Settings: collect General, Backup, Log, Health, Extensions and Files into one
page. The Files tab keeps a panel beside the editor showing what the field
under the cursor does, the file's shape, or the unsaved changes.
- Audit rules: the YAML editor gets the same panel, plus a Test tab that runs a
rule's regex against pasted lines.
- Skills and Agents become separate pages; install, search, updates and trash
move into them. Old routes redirect.
- Tour: rewrite every step's copy, add MCP and Install steps, and anchor each
step to a control that is always rendered so no step silently times out.
- Give every button a pointer cursor; the browser default does not say a button
is clickable and each component class had to opt in by hand.
Docs follow the new Settings tabs and the MCP list.
* chore(devcontainer): move dev servers to uncommon ports
5173, 19420 and 8888 are defaults that other projects on the same
machine already hold, so `make devc` failed to bind and the container
never started. Use 45173, 49420 and 48888 instead, identical inside and
outside the container so every banner, log and doc prints the URL that
actually works.
* feat(ui): link dashboard source counts to their pages
The counts in the playful dashboard's source note named a number but led
nowhere, so reaching Skills or MCP from there meant going back to the
sidebar. Each row is now a link to its own page.
The hover fill uses currentColor because .ss-pinnote.src overrides the ink
tokens for the yellow note.
* feat(ui): build and browse hubs from the dashboard
A hub is a JSON index that lists skills, and until now the web UI could
only consume one someone else had already hosted. Writing the index by
hand was the only way to produce one, and the two halves never met: the
dialog that searched hubs and the page that would have built one shared a
config file but nothing else.
The new /hubs page holds both halves behind two tabs:
Browse pick a hub, filter its contents by tag or text, install from it
without leaving the page
My hubs assemble an index from installed skills, validate it, export it
A draft is offered as a source in Browse straight away, read from the
draft API rather than a file, so you can see what a recipient sees before
deciding whether to host anything. Where the index is hosted is stored on
the draft and stripped from the exported document, so a reopened hub
remembers where it was published and the published file stays clean.
Hub management moves out of the install dialog, which keeps only the
search-source dropdown and links here.
Routing switches to createBrowserRouter for useBlocker, which guards the
builder against navigating away from unsaved entries.
* docs(extras): add shared instructions recipe for AGENTS.md across agents
Answers a recurring question about sharing one instruction file between AI
agents: extras already covers it, since each agent keeps its user-level
copy in its own directory and one extra can target all of them.
Naming the source file AGENTS.md rather than something neutral matters —
Codex concatenates AGENTS.md files and has no import syntax, so it only
recognises the file by name. Claude Code and Antigravity read a fixed
filename of their own but resolve @-imports, so a one-line CLAUDE.md or
GEMINI.md picks up the synced file.
Says plainly that this is not agent memory. Agents keep their own
learnings in private stores — a Markdown directory for Claude Code, a
database for Codex, non-file storage for Cursor — so that side is a
migration problem, not something syncing files can solve.
* feat(ui): filter skills and agents by target
The Targets column showed which targets a resource reaches, but there was
no way to narrow the list down to one. The filter reads the sync matrix —
the same source that column renders — so a result always matches the icons
the rows show.
Options and counts are derived from the matrix, so no target is offered
that would filter to nothing, and the control stays hidden until at least
two targets have something. A selected target that disappears on a kind
switch or an uninstall falls back to All rather than emptying the list.
The toolbar had 24px to spare at 1440px, so the new control wrapped the
row. Source moves from a 325px segmented control to a dropdown like its
neighbours, and every filter now sizes to its label instead of a fixed
width that truncated the longer values.
* fix(ui): tighten page headers and reach hubs from skills
PageHeader carried mb-7 while .ss-wrap already separates its children with a
28px flex gap. Margins and flex gaps add up rather than collapse, so every
page inside the wrapper opened with a 56px hole under its title. Move the
default onto .ss-pgh so the wrapper can zero it, keeping the 28px for pages
that lay out their own children.
Hubs is a source you install from and a catalog you publish, not inventory
you own, so it no longer sits in the Library group next to Skills, Agents and
Plugins. The Skills page header already links to it and the page already
carries a back link there. Skills stays highlighted while the hub page is
open. Plugins takes the Package icon its own page identifies with, so the
sidebar no longer shows two identical puzzle pieces.
* fix(skill): stop implement-feature from running --no-tui as a command
The dispatch-order note wrote "not --no-tui" as !`--no-tui`. In a skill file
that is the syntax for running a shell command at load time, so the skill died
with "command not found: --no-tui" before any of it could be read. Spell the
condition out in words instead.
* feat(list): toggle model invocation for a skill with M
The only cheap gesture in the list TUI was E, which hides a skill from every
harness. Wanting a skill installed but not auto-loaded meant leaving the TUI
to edit SKILL.md, so people disabled skills they only meant to make
manual-only. M flips disable-model-invocation in place and the detail panel
shows a "manual only" badge so the state is never a blind toggle.
Unlike E this edits the skill file, which upstream may own: update skips
tracked repos with local changes, and reinstalling a skill drops the edit. M
says so and asks before writing in those cases. The flag is written as a
single-line edit rather than a YAML round-trip, and turning it back off
removes the line, so two presses return the file byte for byte and a tracked
repo is clean again. Agents are left alone; their harnesses do not read the key.
Refs #283
* feat(plugin): manage complete plugins across agent tools
Add CLI, TUI, and web flows for discovery, adoption, sync selection, updates, and removal while preserving native package formats and unrelated installations.
Support Claude, Codex, Cursor, Antigravity desktop, Pi, and OpenCode. Update help, completions, built-in skill, and documentation; keep plugins outside sync --all.
* docs(skill): describe the redesigned dashboard in the style skill
The style skill still documented the pre-redesign hand-drawn system, so
pages written with it came out on legacy tokens and outside the ss-*
classes; the new Plugins page was the first casualty.
The rewrite names tokens and classes and points to components.css for
their values, because copied values are what went stale last time. The
ui command docs described Playful by the old look as well.
* fix(ui): keep Playful ghost buttons free of stray shadows
The hover and disabled rules outranked the ghost variant's
box-shadow: none, so a frameless button grew a lone shadow arc in
Playful. Bulk toolbar and toast buttons shared the problem.
Ghost also gets a soft outline in Playful: beside 2px ink frames a bare
label such as Cancel did not read as a button.
* refactor(ui): lay out the Plugins page like the MCP page
The page was built on legacy tokens and plain markup, so it missed the
Playful frames and drifted from every other page. A per-agent column
layout would not have scaled either: each agent that gains plugin
support would add a column.
Plugins now follow the MCP server list: one row per plugin with its
agents as toggles inside the row, so the list only grows downwards.
Per-agent sync and update move to package-level actions in the row
menu, and the dialogs use the standard header, body and footer.
* feat(ui): give plugins a kind colour and per-agent actions
Plugins borrowed a neutral grey tile because no kind colour existed,
which made them the one resource without an identity on the page.
Magenta sits furthest from the five hues already taken.
Laying plugins out as rows had dropped the per-agent sync and update.
Retrying one agent after a partial failure should not re-run the rest,
so each agent in the expanded row gets its own menu. Update is left out
for a deselected agent, which has nothing installed to update.
* feat(ui): count plugins on the Dashboard
The source summary listed every resource kind except plugins, so the
Dashboard under-reported what a source holds. The Playful source note
grows to fit a fifth row.
MCP now sits above Plugins in the sidebar to match that order, and its
Dashboard label reuses the sidebar's "MCP" instead of a longer string
of its own.
* feat(scripts): add a host verification environment
Plugin and MCP work talks to the real Agent CLIs, and those live on the
host, not in the devcontainer, where every target reports "not installed".
Verifying by hand meant either pointing a build at real config or
assembling a dozen environment variables each time.
`make verify` builds a host binary and runs it against .verify-home/ in
the repo, redirecting HOME, the XDG dirs and each agent's own config
variable. The environment starts from `env -i`, so a stray
SKILLSHARE_CONFIG in the caller's shell cannot pull the run back to real
paths. Living inside the repo is safe because project-mode detection
only inspects the working directory and never walks up to a parent.
It also seeds the dashboard cache from ui/dist. A dev build serves a
placeholder rather than the working tree, because runtime download only
applies to numbered versions, so the UI was previously unverifiable
without a release.
* fix(ui): show progress while a plugin selection is applied
Toggling an agent shells out to that agent's own CLI, which takes
seconds. The row gave no sign of it: the toggle only greyed out, and
the confirmation badge is hidden while a target is deselected, so
turning one on looked like nothing happened at all.
The toggle being applied now lifts out of the greyed state and spins in
its badge. Agent versions also get room for their real strings, since
"2.1.276 (Claude Code)" was truncated to a width chosen for a bare
version number.
verify.sh gains a 'run' subcommand, so the same throwaway environment
can be driven from a script rather than only an interactive shell.
* fix(ui): spin the control that started a plugin action
Reading the inventory shells out to six agent CLIs and takes about two
and a half seconds, and every action refreshes it afterwards. Nothing on
the page said so: a menu action closed the menu and left the page still,
so the wait read as a freeze.
Each control now tracks its own work. Check, sync and import spin in
place, a menu action spins the row it acts on, and the agents refresh
follows the query itself rather than the mutation flag.
* feat(ui): split MCP add and import by where the configuration comes from
"Import from a target" opened a dialog whose second tab pasted a snippet,
which is not a target and usually comes from a provider's setup page rather
than a tool the user already configured. Pasting is a way to add a server,
so it moves under "Add server" beside the form, and importing now means one
thing: read what an installed Agent already has.
The paste side gains a file picker, matching `mcp import --file`; the API
only ever took content, so the browser reads the file and fills the editor.
The target picker no longer waits for a successful parse, and a footer that
said "writes 0 config files" next to a greyed-out button now names what is
missing instead.
* fix(ui): label the plugins result strip as the last action
The strip and the plugin inventory below it used the same list styling with
no heading on either, so two rows carrying the same plugin name read as one
list. The status word made it worse: "imported" is one of the outcomes
lifecycle.go returns for an action, but it sat where an inventory row shows
state, so it looked like a property of the plugin.
Give it the section heading the Agents list already has, plus a count and a
dismiss button, so it reads as the report it is.
* feat(plugins): expand compatibility and clarify setup guidance
Support additional native targets, safe source links, Git refs, and explicit OpenCode entries while exposing per-target capabilities across CLI, TUI, and UI.
Separate unsupported integrations from environment errors, explain next steps in plain language, and link official documentation. Update built-in guidance, website docs, completions, and regression coverage.
* feat(plugin): serve a plugin's files and translatable discovery problems
The dashboard could only show a plugin's name and targets, so reviewing
what a plugin ships meant opening the snapshot by hand. Files and
ReadFile expose the reviewed local copy read-only through os.OpenRoot,
which keeps reads inside the snapshot even across symlinks.
Discovery problems now carry a key and arguments next to the English
sentence, so the dashboard can translate them while the CLI keeps its
wording. Inventory asks every Agent CLI in parallel and Packages answers
from config alone, so the plugin list no longer waits on the slowest CLI.
* feat(mcp): sync to Pi, show native config, and lay out written entries
Pi has no built-in MCP support, so a server targeting it must name the
third-party extension that will read the file. `--pi-extension` and the
`piExtension` source field record that choice, and sync writes the file
format of the chosen package instead of guessing.
RenderNative and POST /api/mcp/render return what each Agent would
receive for a definition, with secrets kept as references, so the
dashboard can show the result before anything is written.
Sync used to insert entries on a single line and never revisit them. A
managed entry that still sits on one line is now reported as an update
and rewritten one field per line; entries formatted by hand are left
alone.
* feat(ui): choose plugin Agents like MCP, browse plugin files, view MCP config
The plugin fold listed bound Agents with a per-Agent menu, while MCP let
you simply tick the Agents you want. Plugins now work the same way: the
fold discovers where the source can also go, shows those Agents as
unticked toggles, and a tick previews an install. The per-Agent menu is
gone, and updating a plugin targets only the Agents that support it
instead of hiding the action when one of them does not.
A plugin with a local copy can be browsed in a read-only files dialog,
using the same tree as a skill's Files tab. The MCP "view config" moved
out of a fold in the server form into its own view, reachable from the
server menu, so the native result has room to be read. The MCP dialogs
also ask for the Pi extension when Pi is selected.
Playful tile names drop from 25px to 22px: "brainstorming" was 1px wider
than the tile and lost its last letter to the next line.
* docs: reshoot dashboard screenshots for the redesigned UI
The README and website still showed the pre-redesign dashboard, which no longer matches what users see after upgrading.
* docs: describe MCP, plugins and the new dashboard for v0.21.0
The README still announced v0.20.0 and never mentioned plugins, and its
MCP section listed four clients. The built-in skill and the docs pick up
the Pi extension setup, the formatting-only MCP update, and the plugin
fold and files dialog, so agents and users are not surprised by them.
Skills from non-GitHub git sources (GitLab, Gitea, self-hosted, SSH)
render a Remote badge but matched no filter chip, so they were only
reachable through All. The Local chip had the same gap for skills whose
metadata type is "local" rather than empty.
Source filters now classify skills with the badge's own resolver instead
of re-implementing it per page, and FilterType is derived from SourceType
so the two cannot drift apart again. Applies to the resources, batch
uninstall and update pages.
Refs #278
SourceBadge only recognised github/github-subdir metadata types, so skills
installed from Gitea, GitLab, self-hosted or SSH git sources fell through
to the Local badge. Add a Remote badge for any non-local install source,
matching the CLI list view.
Refs #270
Hub fetch failures surfaced only a bare "fetch hub: HTTP 400" toast, with no indication of which hub failed or how to fix it.
Hub load errors now name the likely cause and include the URL for every failure mode (4xx/auth, malformed URL, invalid JSON, unreachable host, missing local file). The Search page renders them as a structured toast titled with the failing hub, and Toast gained an optional title for this.
The web UI skill preview always fetched SKILL.md via api.github.com, so skills from GHE or other non-github.com hubs showed only index metadata (name, description, tags) instead of the full rendered content.
FetchPreview now resolves the source platform: github.com and GHE use the host's REST Contents API, while other platforms (GitLab, Bitbucket, Azure, SSH) fall back to a shallow clone. Sources whose content can't be retrieved degrade gracefully with a clear notice in the modal.
Fixes#193
- Select: render the dropdown through a portal so it escapes ancestor
stacking contexts (it was clipped behind the search box); in the
playful theme match the trigger's wobble radius instead of the inline
uniform radius
- devcontainer: mount ui/ and website/ node_modules as anonymous volumes
so platform-native rolldown bindings install in-container (Vite failed
to start with host darwin bindings); force CHOKIDAR_USEPOLLING because
virtiofs does not reliably deliver inotify events, so saves were
silently missed and HMR stopped firing
- SearchPage: remove the "submit a PR to skillshare-hub" prompt
- i18n: drop the now-unused search.hub.submitPrPrompt key and add the
missing extras.* keys across all locales to restore key/placeholder
parity with English
UI (Extras page):
- Label per-target flatten/extension/mode controls with column headers
- Render flatten as a dropdown so it matches the adjacent selects
- Collapse the add-target form behind a button and widen the path input
- Confirm before removing a target
- Polish layout: tinted icon, inset source chip, hover rows, aligned columns
Components:
- Select: bound the dropdown width so long option descriptions wrap instead
of stretching the popup across the page
- Toast: render messages multi-line, widen the container, and give
error/warning toasts more time to read
Sync feedback:
- Restructure extras sync-failure toasts into readable lines and fix the
contradictory "N more errors" count (new nMoreErrors translation)
CLI/server:
- Match extra target paths by resolved absolute path (handles ~ vs /home)
when adding, removing, or updating targets
The mixed-tracked-repo recovery path used a fragile substring match on
the error message and re-cloned the repo via /discover to count skills
and agents. Both problems share one root cause: the counts InferTrackedKind
already computed during the failed first install were thrown away.
- Add install.TrackKindAmbiguousError carrying the discovered Skills /
Agents counts. InferTrackedKind returns it on the ambiguous branch.
Error() preserves the legacy message verbatim so CLI output is unchanged.
- Server handleInstall track branch uses errors.As to detect it and emits
writeCodedError with stable code 'install.track_kind_ambiguous' plus
skills/agents counts in error_params.
- UI replaces the substring match with an ApiError.code check and reads
counts straight from err.params, skipping the second /discover clone.
- Add unit tests asserting errors.As, count fidelity, and explicit-kind
short-circuit (no clone).
Net result: 2 clones instead of 3 on the mixed-track recovery path, and a
stable cross-language contract instead of a brittle English substring.
Installing a mixed git repo (containing both skills and agents) with
'Track' enabled failed with 'tracked install is ambiguous; pass --kind
skill or --kind agent', and the UI gave no way to recover.
- api.install: add optional 'kind' param (backend already supports it)
- InstallForm: detect the ambiguity error, fetch counts via /discover,
and prompt the user to pick Skills or Agents via a kind-selector
dialog; re-run install with the chosen kind
- Snapshot all install params (source/name/into/branch/force/skipAudit)
at error time so a later kind selection cannot drift if the user
edits the form behind the dialog
- Fall back to the original ambiguity error if /discover fails, so the
user is not left with only a confusing secondary error
- Render the kind selector in both collapsible and non-collapsible
return branches so the fix works for any InstallForm caller
- preventClose while installing so an in-flight kind selection is not
cancelled by Escape / backdrop click
- Hoist KindOption out of the render body to keep its component
identity stable across parent re-renders (no more autoFocus refire)
UI redesign of the kind selector:
- Replace ConfirmDialog wrapper with DialogShell + stacked card-style
kind buttons (icon + label + count + chevron)
- Title row gains an X close affordance; remove the redundant bottom
Cancel/divider to match other modals
i18n: drop legacy skillsButton/agentsButton/cancelText keys; add
skillsLabel/agentsLabel/itemCount across all 11 locales with fully
native phrasing (no mixed English in non-English locales).
Three user-visible additions to the web dashboard, plus the CLI surface
needed to drive them.
* `skillshare ui start` / `skillshare ui stop` — run the dashboard as a
managed background process. Re-running start reuses the existing
healthy process; stop uses the remembered host/port. Foreground
`skillshare ui` keeps working unchanged. Process tracking lives in
ui_process_{unix,windows,other}.go via a PID + current-state file.
* `--app` — opens the dashboard in a Chromium app-mode window (Chrome,
Edge, Brave) on macOS, Windows, and Linux for a separate Dock/taskbar
entry without browser chrome.
* In-place upgrade from the browser — the Update dialog and the Doctor
page's Version card now expose an Update now button. Backed by two
new endpoints (POST /api/upgrade, POST /api/restart in
handler_restart_upgrade.go); the browser polls GET /api/health and
reloads once the new server is ready. Dev builds short-circuit with
`devMode: true` and a simulated restart so the flow can be exercised
without rebuilding the binary.
* Progressive Web App — site.webmanifest, app icons (192/512), and a
minimal service worker registered from main.tsx. Supporting browsers
can install the dashboard as a standalone desktop app.
* Doctor Version card redesign — state-coloured icon (blue when an
update is available, green when up to date), explicit
`current → latest` delta, Update button only when there is something
to upgrade.
* i18n — translate "DEV mode" in zh-TW / zh-CN / ko (previously left
as raw English).
Adds CHANGELOG.md and website changelog entries for v0.19.20 and
refreshes website/docs/reference/commands/ui.md to document the new
modes, --app flag, new REST endpoints, and the in-place upgrade flow.
- Add ErrSkillNotFound sentinel error in search/preview.go so callers
can distinguish 404 from other errors with errors.Is instead of
brittle string matching
- Cap previewCache at 200 entries with TTL purge + full reset on
overflow to prevent unbounded memory growth
- Make Card and SearchResultsTable rows keyboard-accessible
(role=button, tabIndex=0, Enter/Space handlers)
- Refactor SkillPreviewModal to accept a single SearchResult prop
instead of 6 individual fields — reduces prop drilling and keeps
the modal in sync with the result shape
- Fix search_online_test: add --list flag so output is not gated
behind interactive TUI
Add a preview modal so users can read a skill's SKILL.md content
before installing. Clicking a search result card or table row opens
the modal; the Install button is now inside the modal.
Backend:
- New GET /api/preview?source=owner/repo/path endpoint
- FetchSkillContent fetches SKILL.md from GitHub Contents API
- Git Tree API fallback for hub sources where the install shorthand
doesn't match the repo's actual file path
- In-memory TTL cache (5 min) via sync.Map
- Proper 404/429/500 error handling with rate-limit support
Frontend:
- SkillPreviewModal component (DialogShell-based, lazy-fetch)
- Cards and table rows clickable to open preview
- Graceful fallback: if fetch fails but hub metadata exists, show
description + tags instead of error
- Hide Stars column in table view for hub mode
- Fix Card component missing onClick prop passthrough
- Fix IntersectionObserver not reconnecting after loading more items
- Prevent background scroll when any modal is open (DialogShell)
- i18n keys added to all 11 locales
Closes#141
Add PATCH /api/resources/{name}/source endpoint to update the source URL
for installed skills. For tracked repos, all sibling skills sharing the
same git remote are updated together with the new repo base URL, and the
git remote origin is also updated.
Frontend changes:
- Add Source URL input field in the skill editor (visible for all skills
with metadata)
- Use shared Input/Textarea components in FrontmatterEditor for visual
consistency, with new size='sm' variant
- Replace custom kind-badge/tracked-badge with shared KindBadge/SourceBadge
components in the editor header
- Preserve original YAML frontmatter formatting when only the body or
source URL is changed (avoids spurious quote style / line wrap diffs)
- Skip diff review dialog when only source URL is modified
Backend fixes:
- Two-pass skill lookup in handleGetSkill: prefer exact FlatName match
over baseName fallback to prevent name collisions between standalone
and tracked repo skills
- Add basename-only fallback in MetadataStore.GetByPath for tracked repo
skills stored with short keys
- Use cfg.Source (global) consistently in save/patch handlers to match
the source used by list/detail handlers in project mode
Closes#142
Extract highlightArgs utility from SkillEditor to shared lib so both
the editor and ResourceDetailPage render $ARGUMENTS with the warning
badge style.
- Add p/li component overrides in ResourceDetailPage mdComponents
- Move .arg-token CSS from scoped editor styles to global index.css
- Simplify argsCount to hasArgs boolean (count display removed)
- Remove ×{count} prefix from argsHint in all 11 locales
The shared SegmentedControl has no gap between icon and label, making
Edit/Split/Preview look cramped. Revert the visual to the original
editor-themed .seg-btn/.seg-group CSS (4px icon gap, paper background,
shadow-sm active state).
Keep the DRY JSX win by introducing a small EditorSegment helper that
renders the same markup the three inline call sites previously duplicated.
Follow-up to the earlier simplify pass — addresses the larger refactors
called out by the three review agents:
- Drop local SegmentedField in favour of the shared SegmentedControl used
across the dashboard; reuse it for the Fields/YAML toggle, the
Edit/Split/Preview mode switcher, and the enum frontmatter fields. Removes
~34 lines of duplicated .seg-group / .seg-btn CSS.
- Split the 80-line FrontmatterField if/else into per-type subcomponents
(EnumField / BoolField / ArrayField / TextField) dispatched from a small
FieldControl switch. The synthetic 'conditional' field type is gone —
visibility is now a plain 'showWhen' predicate on a 'text' field, with an
optional 'placeholder' override.
- Extract a shared CollapsibleGroup wrapper; both FrontmatterGroup and
FrontmatterMetadataGroup use it, deleting the copy-pasted fm-group chrome.
- requestAnimationFrame-throttle the synced-scroll handlers. Previously
every onScroll event (~60×/s) ran buildSyncAnchors, which measures
scrollHeight / getComputedStyle / querySelectorAll and sometimes the DOM
mirror. Now one pass per frame per scroll source, with RAF + timer
cleanup on unmount.
useLineDiff consolidation was considered but skipped — computeLineDiff
returns only add/remove rows, whereas DiffView needs the eq rows for the
side-by-side view, so merging would complicate the shared hook.
- Drop unused EditorCustomField / initialCustomFields / draftTargets / FileText
re-export; narrow onSaved signature to (nextContent: string).
- Gate DiffView render on showDiff so the O(n·m) LCS + composeSkillMarkdown
don't run on every keystroke while the dialog is closed.
- Route the markdown preview and Outline through useDeferredValue so typing
stays responsive while react-markdown catches up.
- Serialize frontmatter YAML lazily (only when yamlMode is active).
- Equality-guard the metadata-row reconcile effect to avoid no-op setState
storms on every frontmatter keystroke.
- Memoize derived stats (tokens/words/lines) into a single pass instead of
two splits per render.
- Replace the private YAML parser in ResourceDetailPage with
lib/frontmatter's parseSkillMarkdown; removes ~60 LOC of duplicate logic.
- Remove WHAT-narrating comments ("Keyboard shortcuts", "Derived stats",
"Stage 1: Content Stats Bar", etc.).
Introduces a full in-browser skill editor reachable from the skill detail
page, replacing the previous read-only detail view for edits.
Editor
- Two-pane layout with Markdown textarea, live preview, and synced scroll
- Edit / Split / Preview mode toggle in the status bar (⌘P to cycle)
- Keyboard shortcuts: ⌘S save (with diff review), Esc cancel
- Outline drawer with jump-to-heading
- Token / word / line / file count in status bar, 5K budget warning
Frontmatter
- Grouped field editor (Identity / Invocation / Execution / Metadata)
matching the 13 official SKILL.md fields
- Shared 1,536-char budget enforced across description + when_to_use
- Switch toggles for booleans, segmented control for enums, chip array
inputs for list-valued fields
- Metadata group edits nested metadata.* keys directly, with stable row
ids so typing key → clicking value no longer remounts the inputs
- Special handling for metadata.targets so values serialize as a YAML
list (backend ParseFrontmatterList requires arrays)
- YAML mode for raw editing; Fields ↔ YAML round-trips cleanly
- Legacy root-level 'targets:' migrated to metadata.targets on load
Diff review
- Save triggers a side-by-side diff modal built on the shared DialogShell
with focus trap and backdrop, using the Button component for actions
- YAML serializer emits plain scalars when safe instead of wrapping every
value containing ':', '*', '#', or '"', eliminating spurious churn in
the save diff
Misc
- DialogShell gains 4xl–7xl maxWidth presets
- Resource detail sidebar shows a Targets row when metadata.targets is set
Backend additions (server handlers):
- POST /api/resources/:name/content save SKILL.md content
- POST /api/resources/:name/open open in local editor
UpdatePage uses variant="warning" on the Force Retry button for
blocked items to visually distinguish them from error-state Force
Retry (which uses danger). The Button component didn't expose a
warning variant, causing the Docker build to fail with a TS2322
type error.
Adds warning alongside existing danger variant, mirroring the same
outlined-with-hover-fill pattern and matching the warning color
scheme already used by Badge and in-line text-warning classes.
Add GET/PUT /api/agentignore endpoints mirroring the existing
skillignore API, backed by skillignore.ReadAgentIgnoreMatcher()
and resource.AgentKind{}.Discover() for ignore stats.
On the frontend, extend the Config page with a third tab for
editing .agentignore. Generalize the former SkillignoreTab into
a shared IgnoreTab component with a kind prop, and extend
AssistantPanel to show ignored agents in the sidebar.
Agent discovery was treating non-agent files (CLAUDE.md, AGENTS.md, etc.)
as agent resources when they appeared in tracked repos.
Two fixes:
- Expand ConventionalExcludes with 7 well-known convention filenames
(CLAUDE.md, AGENTS.md, GEMINI.md, COPILOT.md, CONTRIBUTING.md,
CODE_OF_CONDUCT.md, SUPPORT.md)
- For tracked repos with an agents/ subdirectory, only discover files
inside agents/ — skip repo root files and other subdirs (docs/, etc.)
Also includes UI polish:
- Display flat names (__ separator) as path separators in toasts/labels
- Consolidate batch install errors into single toast
- Show .agentignore state in sync diff/preview responses
- Track badge + friendly repo name in resource list/folder tree
- auditCache: centralized cache with installed-count staleness check
- resourceNames: agent display name formatting (strip .md suffix)
- AuditPage: skip audit when 0 installed, show empty state per kind
- clearAuditCache on install/uninstall/update across all pages
- ResourcesPage: agent target summary, context menu for agents
- FilterStudioPage: agent display name in preview search
- SkillPickerModal: accessible radio button (button element)
- fieldDocs: agents sub-keys no longer marked reserved
- Integration tests: diff/sync prune after uninstall-all, tracked agent install
Add local-path resource discovery so install flows can detect agents as well as skills. Preserve nested agent paths and suffixless agent metadata keys during batch install and update, and refresh the UI copy to use resource-aware wording on the install surface.
- Add agent-mode batch uninstall handler with kind-based routing
in handler_uninstall.go (skill/agent dispatch)
- Add Skills/Agents tab navigation to BatchUninstallPage and TrashPage
with per-tab filtering and scoped empty states
- Extract SourceBadge component to replace duplicated getTypeLabel /
skillTypeLabel helpers across ResourcesPage, ResourceDetailPage,
BatchUninstallPage, and AnalyzePage
- Add kind field to BatchUninstallRequest API type
- Capitalize badge labels for consistency (tracked → Tracked, etc.)
Backend:
- PATCH /api/targets/:name accepts agent_include/agent_exclude fields
- POST /api/sync-matrix/preview returns agent entries with kind=agent
- GET /api/overview returns agentsSource and extrasSource paths
- Fix ResolveProjectTargets to resolve Agents sub-key (path + filters)
- Add agent filter CLI flags: --add-agent-include, --add-agent-exclude,
--remove-agent-include, --remove-agent-exclude, --agent-mode
- Unify target list JSON/TUI to show agent info and filters
Frontend:
- FilterStudioPage: kind-driven single-context view (?kind=skill|agent)
- KindBadge: use design system tokens (ss-badge + wobble in playful)
- TargetsPage: link to Filter Studio with ?kind= param
- ResourceDetailPage: capitalize type labels (Local, GitHub)
- ResourcesPage: sync activeTab from URL searchParams on navigation
- DashboardPage: show agents/extras source dirs, link skills card
with ?tab=skills for reliable tab switching
- API client: extend updateTarget and previewSyncMatrix for agent filters
Tests:
- Agent include/exclude round-trip persistence
- Invalid agent pattern returns 400
- Preview with agent entries and kind field
- No agent entries for targets without agent path
- CLI target agent filter integration tests
Add Config.HasAgentTarget() helper that checks if any configured target
has an agents path set.
handler_install: append warning to install summary when agents are
installed but no targets support them.
handler_sync: after sync completes, check if agents exist in source
but no targets have agent paths — add warning to response.
SkillPickerModal: use items-start + mt-1.5 on checkbox/radio to align
with first line of multi-line content (name + path).
handler_install: summary message uses 'agents' when kind is agent,
instead of always saying 'skills'.
Backend (install_discovery.go):
- Add discoverAgents() call to all 3 subdir discovery paths (sparse
checkout, GitHub API, full clone fallback). Previously only whole-repo
discovery detected agents.
Frontend (InstallForm.tsx):
- Single-skill fast path now checks hasAgents before auto-installing.
When agents exist alongside a single skill, falls through to the
kind selector instead of silently installing as skill.
Backend (handler_install.go):
- handleInstallBatch checks body.Kind field; when 'agent', uses
InstallAgentFromDiscovery() writing to agentsSource() instead of
skills source directory.
Frontend (client.ts):
- installBatch accepts optional kind param.
Frontend (InstallForm.tsx):
- handleBatchInstall detects kind from selected items and passes
kind='agent' to the API when installing agents.
Backend (handler_install.go):
- handleDiscover now returns agents[] alongside skills[] with kind field
- needsSelection true when agents are present
Frontend (client.ts):
- Add DiscoveredAgent type and agents field to DiscoverResult
Frontend (InstallForm.tsx):
- Mixed repo: show kind selector dialog (Skills/Agents buttons)
- Pure agent repo: map agents to skill picker format directly
- After kind selection, open existing SkillPickerModal with filtered items
Spec/design/tasks updated with kind-first selection decision.
Rename SkillsPage.tsx to ResourcesPage.tsx. Page title changed to
'Resources' with subtitle showing both skill and agent counts.
Add top-level SegmentedControl tabs (Skills | Agents) for switching
between resource types. Each tab has its own count badge. Sub-filters
(All/Tracked/GitHub/Local) apply within the active tab.
Tab selection persists via ?tab=agents URL param. Dashboard stat cards
link to /resources and /resources?tab=agents.
Route: /resources is the primary path; /skills kept as alias.
Nav sidebar: 'Skills' → 'Resources' with Layers icon.
Keyboard shortcut 's' → /resources.
Tour updated for resource-oriented language.
list_tui_item: show [A] badge prefix for agents in terminal list view.
list.go: add Kind field to skillEntry and skillJSON structs.
SyncPage: add kind badge to diff items using DiffTarget.items.kind.
AuditPage: add kind badge to audit result cards.
UpdatePage: add kind badge to update result items.
SkillPickerModal: add kind badge in skill picker.
client.ts: add kind to DiffTarget items and AuditResult types.
Switch DialogShell backdrop detection from onClick to onMouseDown.
The click event's target resolves to the nearest common ancestor of
mousedown/mouseup positions, so dragging a text selection from an
input onto the backdrop would fire onClose. mousedown's target is
always the element directly pressed, so only an intentional press on
the backdrop itself triggers close.
- Add explicit undefined initializer to useRef<T>() for React 19 compat
- Remove unused isProjectMode from DashboardPage (dead after refactor)
- Remove unused queryClient/toast from SkillsTable (moved to useSkillActions)