Add config.BaseDir() as single source of truth for base directory
resolution. Replace all hardcoded ~/.config/skillshare paths.
Path resolution priority:
1. SKILLSHARE_CONFIG env var (unchanged)
2. $XDG_CONFIG_HOME/skillshare (any platform)
3. %AppData%/skillshare (Windows only)
4. ~/.config/skillshare (Linux, macOS fallback)
- Add config.BaseDir() with XDG_CONFIG_HOME + Windows %AppData% support
- Update BackupDir(), TrashDir(), init default source to use BaseDir()
- Show resolved base directory in doctor output
- Add unit tests and integration tests for XDG override
- Document XDG_CONFIG_HOME and update Windows paths in docs
Project targets now use the same short names as global targets (e.g.,
"claude" instead of "claude-code"). Old project_name values are preserved
as aliases so existing configs continue to work.
- Add Aliases field to targetSpec (deprecated, remove once safe)
- Update 11 targets in targets.yaml: project_name = global_name
- LookupProjectTarget falls back to alias lookup
- Fix 3 callers that bypassed LookupProjectTarget
- Extend MatchesTargetName and KnownTargetNames with alias support
- Add TestLookupProjectTarget_Alias
- Update all integration tests, docs, CLI help, and FAQ
- Update Homebrew command from tap (runkids/tap/skillshare) to
homebrew-core (brew install skillshare) in README, website homepage,
and getting-started docs
- Add note that Homebrew build is CLI-only (no web dashboard)
- Add IsUIBuilt() detection in embed.go/embed_dev.go to check for
placeholder UI at startup
- Show CLI warning with install instructions when skillshare ui is
run on a build without the frontend
- Add Homebrew Note section to commands/ui.md with migration steps
Support -p/--project flag with auto-detection, matching other
dual-mode commands. Show include/exclude filters in diff output
so users understand why certain skills are absent from results.
Skip git-not-initialized warning in project mode since the project
itself has version control and backups aren't used. Skip tracked repo
directories (_prefix) in SKILL.md validity check since they are
container directories for nested skills by design.
Collision checks previously ran on all skills before target filters,
producing false positive warnings when users route same-named skills
to different targets via include/exclude. Now collision detection
runs per-target after filtering, downgrading to info when filters
correctly isolate duplicates.
- Add CheckNameCollisionsForTargets() to internal/sync
- Replace collision blocks in sync.go and sync_project.go
- Show human-readable filter summary per merge target
- Add unit tests (5 cases) and integration test
- Update docs with new two-tier collision output examples
CLI unlinkMergeMode() used exact name matching (sourcePath + entryName)
to identify managed symlinks, which failed for nested skills (flat name
like _team__frontend__ui doesn't match source path _team/frontend/ui)
and orphaned symlinks. Changed to PathHasPrefix-based detection that
checks if a symlink points anywhere under the source directory.
Server handleRemoveTarget() only handled symlink mode (entire dir is a
symlink). Added merge mode branch with unlinkMergeSymlinks() that
iterates target entries and removes symlinks pointing under source.
Hub search was capped at 20 results because it shared the same
normalizeLimit() path as GitHub search. Hub indexes are static JSON
(~100 entries), so returning all results is expected.
- Remove normalizeLimit() from hub search path; limit=0 means no limit
- Backend defaults to limit=0 for hub requests, 20 for GitHub
- CLI uses limit=0 when --hub is set without explicit --limit
- Frontend searchHub() no longer sends a limit parameter
- Replace IIFE rendering with useMemo for stable filter results
- Add IntersectionObserver-based scroll-to-load (20 items per batch)
Hub sources are now saved in config.yaml (both global and project),
replacing the Web UI's localStorage-based approach. This allows CLI
and UI to share the same hub list.
- Add HubConfig/HubEntry types and helper methods (resolve, add, remove)
- Add CLI subcommands: hub add/list/remove/default
- Search --hub now resolves saved labels (e.g. --hub team)
- Add REST API endpoints for hub CRUD (GET/PUT/POST/DELETE /api/hub/saved)
- Migrate SearchPage from localStorage to server API calls
When `skillshare audit <path>` targets an existing file or directory,
config is not needed. Previously config.Load() ran unconditionally,
causing "config not found" errors in CI environments without a
skillshare config. Now path mode bypasses config loading entirely.
- Fuzzy match on skill name, substring match on description and tags
- Display tags (#tag) in both list mode and interactive selector
- Always show "Search again" option when filtering
- Fix --index-url → --hub in hub index help text
- Fix gofmt struct field alignment in SearchResult after adding Tags
- Add missing tags field to frontend SearchResult TypeScript interface
Hub JSON entries can now specify a `skill` field to target a specific
skill within a multi-skill repo. CLI auto-passes -s flag, Web UI
skips the picker modal when skill is specified.
Also adds Community Hub section to search docs with link to
runkids/skillshare-hub.
Rename default output from index.json to skillshare-hub.json for
uniqueness. Add website docs for hub command and hub index guide.
Update search command docs with --index-url section. Add hub and
hub-index to sidebar.
Add `skillshare hub index` subcommand with --full, --source, --output
flags and mode-aware source resolution. Register hub in command map
and usage help.
Integration tests cover: index generation, custom output path, empty
source, full mode, minimal metadata suppression, help text, and
search --index-url with local files (including equals syntax and
browse-all).
SearchFromIndexURL consumes hub-generated index.json (local or HTTP),
with sourcePath resolution, relative path detection (including Windows),
and limit enforcement. The search CLI routes to index search when
--index-url is provided, with conditional spinner text and scoped
GitHub-specific error handling.
- init: replace shallow os.ReadDir count with DiscoverSourceSkills()
to correctly count nested skills by SKILL.md presence
- handler_git: guard against nil slice from GetDirtyFiles() overwriting
the initialized empty slice, which serialized as JSON null
- GitSyncPage: add optional chaining on status.files for defense
- tryPullAfterRemoteSetup: replace git pull/merge with git reset --hard
to avoid merge commits and ff-only/conflict issues when pulling from
remote into a freshly initialized repo
- Add hasGlobalOnlyInitFlags() so init --remote works correctly even
when cwd has .skillshare/config.yaml (project mode auto-detection)
- Update cross-machine-sync doc to reflect reset behavior
Move path display to detection phase output. Multi-select options
now show only name + status instead of name + full path + status,
preventing an unreadable long line when all targets are selected.
Init now creates an initial commit (including .gitignore) so that
push/pull work immediately after setup. When a remote is configured,
init fetches and auto-pulls existing skills from the remote.
- Always create initial commit in doGitInit (fixes stale untracked
.gitignore blocking pull)
- Add ensureGitIdentity: set repo-local fallback when git user.name
is not configured, with hint to set global config
- Add tryPullAfterRemoteSetup: fetch remote after adding origin,
auto-pull if remote has content and local is empty
- Add hintGitRemoteError: context-aware hints for SSH, URL, and
network errors in push/pull
- Show Pro tip only when built-in skill is installed
- Sandbox: add --bare mode (skip auto-init), --volumes for down
(full reset), and sandbox-reset / sandbox:reset targets
- Docs: update init flow diagram, cross-machine-sync guide,
push/pull prerequisites, getting-started, troubleshooting
Allow installing skills into subdirectories for organized folder
structures. Works with all install modes (direct, discovery, tracked)
in both global and project mode.
- Add Into field to InstallOptions and IntoPath() validation
- Update 4 dest-path construction points in CLI with destWithInto()
- Update InstallTrackedRepo for --into prefix in dest and gitignore
- Rewrite ReconcileProjectSkills to use filepath.WalkDir (recursive)
- Add --into to Web UI install handler, API client, and form
- Add 5 integration tests covering local, multi-level, validation,
project mode, and no-source rejection
Previously `skillshare uninstall vue-best-practices` failed when the
skill lived under an organizational folder (e.g. frontend/vue/).
Now both global and project mode fall back to a basename search across
nested directories, matching the behavior already in check/update.
- Add resolveNestedSkillDir() with ambiguity detection
- Use as fallback in resolveUninstallTarget (global) and
cmdUninstallProject (project)
- Also matches _name variant for nested tracked repos
- Add integration tests for basename and full-path uninstall
Multiple agents (Amp, Codex, Copilot, Gemini, Goose, Kimi, OpenCode,
Replit) all map to .agents/skills in project mode. Previously init -p
showed each as a separate selectable entry pointing to the same directory,
which was redundant and confusing.
Now targets sharing a project path are collapsed into a single group entry
(e.g. "agents") with member names shown in parentheses. Discover uses
path-based dedup so an existing "amp" entry correctly prevents the
"agents" group from appearing as new. --select accepts both canonical
and member names for backward compatibility.
- Add project_name/project_path to agents meta-target in targets.yaml
- Add GroupedProjectTargets() in config package with unit tests
- Rewrite detect/list/prompt/discover in init_project.go
- Add agents target to playground demo project
GetTrackedRepos() and GetUpdatableSkills() now walk subdirectories
recursively so skills organized in folders (e.g. frontend/vue/) are
detected by check and update commands. Previously only direct children
of the source directory were scanned.
Also adds:
- Short name resolution for `update` (e.g. `update vue-best-practices`
resolves to `frontend/vue/vue-best-practices`)
- Fix ReadMeta nil check in updateSkillOrRepo to ensure fallback fires
- Remove logo from `check`, replace with mode-aware header
- New docs guide: Organizing Skills with Folders
- Rename screenshot to organizing-skills-comparison.png
Add REST API (GET/PUT/POST /api/audit/rules) for managing custom
audit-rules.yaml through the web dashboard. Extract template and
validation logic from CLI into reusable audit package functions
(ValidateRulesYAML, InitRulesFile, DefaultRulesTemplate). Frontend
provides CodeMirror YAML editor with dirty tracking, file creation
flow, and mode-aware path display. Sidebar nav gains overflow scroll.
Move 17 hardcoded audit rules from Go structs to embedded rules.yaml,
enabling community PRs on rule definitions without recompilation.
Add user-customizable audit rules at two levels:
- Global: ~/.config/skillshare/audit-rules.yaml
- Project: .skillshare/audit-rules.yaml
Merge semantics: built-in → global → project. Users can append new
rules, override existing ones by id, or disable them (enabled: false).
Replace hardcoded localhostURL check with generic exclude regex field
on rules. Add --init-rules flag to scaffold a starter rules file.
Public API (ScanSkill, ScanContent, Finding, Result) unchanged —
callers require zero modifications.
Built-in skillshare skill was auto-installed during `init` and
unconditionally refreshed by `upgrade --skill`. This caused
unintended pushes and surprised new users who didn't need it.
Changes:
- init: add --skill/--no-skill flags, prompt [y/N] (default N)
- upgrade: decouple --skill from --force, prompt when not installed
- Add 4 new integration tests for the opt-in behavior
- Update existing tests to use --no-skill or --force as needed
- Document all non-interactive init flags in website docs
- Add --cmd, --status, --since, --json flags to CLI log command
- Auto-skip irrelevant log section when --cmd narrows to one file
- Add blank line separator between log entries for readability
- API returns totalAll and commands for smarter frontend filtering
- Custom HandSelect dropdown matching hand-drawn design system
- Empty state distinguishes "no matching entries" from "no entries"
- Update website docs and built-in skill reference with new flags
`skillshare version` keeps the ASCII logo, while `-v`/`--version`
now outputs `skillshare v<version>` — the plain-text format that
mise's registry test expects (`skillshare v{{version}}`).
Add registry/skillshare.toml for the upcoming jdx/mise PR.
ProjectSkill gains `tracked: true` (omitempty) so `install -p` can
correctly dispatch tracked repos via InstallTrackedRepo instead of
regular Install. ReconcileProjectSkills now detects .git/ dirs and
reads git remote origin URL for tracked repos that lack meta files,
fixing the bug where tracked repos were silently skipped during
reconcile. SkillName validation is moved after the tracked branch
so _ prefixed repo names no longer fail.
Docs updated to position config.yaml as a portable skill manifest
usable by teams, open source contributors, and community templates.
- Add oplog write helpers to server (oplog_helper.go) and instrument
all handler_*.go endpoints with ops/audit logging
- Improve audit CLI with structured auditRunSummary and severity counts
- Enhance log CLI with richer filtering and detail display
- Improve sync CLI with drift detection summaries in oplog
- Reorder web UI sidebar nav by user workflow (core → acquire → distribute → maintain → monitor)
- Enhance LogPage with better formatting and filtering
- Add server oplog unit tests (logging_test.go, log_test.go)
- Update docs for log command and quick-reference
Sync discovery (DiscoverSourceSkills) was skipping all hidden directories
via utils.IsHidden(), causing skills inside .curated/ and .system/ (like
openai/skills) to be invisible after install. Now only .git is skipped,
matching the install-time discovery that was already fixed in 8ec5306.
Tracked repo naming now uses owner-repo format (e.g., _openai-skills
instead of _skills) via new Source.TrackName() method, preventing
collisions between repos that share the same basename.
Project-mode `list` command now uses recursive DiscoverSourceSkills()
instead of shallow os.ReadDir(), aligning with global list and web UI
behavior so tracked repo skills are visible everywhere.
Also includes CHANGELOG for v0.11.0, README updates, docs improvements,
and minor UI nav reorder.
Add oplog package (internal/oplog) that records every mutating command
(install, uninstall, sync, push, pull, collect, backup, restore, update,
audit) as a JSONL entry with timestamp, status, duration, and args.
- CLI: `skillshare log` with --audit, --tail, --clear, --project flags
- API: GET/DELETE /api/log endpoints with type and limit params
- Web UI: LogPage with ops/audit tabs, table view, clear and refresh
- Instrumentation across 12 command handlers (CLI + server)
discoverSkills() skipped all dot-prefixed dirs, causing repos like
openai/skills (which use .curated/ and .system/) to find 0 skills.
Fix: only skip .git, traverse all other hidden dirs.
CLI --skill flag now falls back to fuzzy matching (via lithammer/fuzzysearch)
when exact match fails. Single fuzzy match auto-selects; ambiguous matches
show "did you mean" suggestions.
UI SkillPickerModal gains a filter input (autofocus, substring on name+path)
that appears for >5 skills. Select All toggles only filtered skills.
Batch install toast fix: always show summary toast even when some skills
are blocked by security audit (previously early-returned with no feedback).
Audit dialog now targets only blocked skills for force-retry.
CLI output for mixed results: use warning banner instead of success,
show failures first, condense >10 successes into a single line.
Detect when targets have fewer linked skills than source and surface
warnings across CLI (status, doctor) and web UI (Dashboard, Targets).
Helps users notice unsynced skills after installing new ones.
- Add Web UI Trash page with list, restore, delete, and empty actions
- Add 4 server handlers (handler_trash.go) with dual-mode support
- Fix handleUninstallSkill/handleUninstallRepo to use trash.MoveToTrash
instead of os.RemoveAll (bug fix: uninstall via UI now soft-deletes)
- Add trash delete/empty CLI subcommands with integration tests
- Show Uninstall button for all skills in SkillDetailPage (not just
tracked/installed ones)
- Update confirm dialog wording to reflect soft-delete behavior
- Update docs with new trash subcommands
Replace permanent deletion with soft-delete: uninstall now moves skills
to a trash directory instead of using os.RemoveAll. Skills are retained
for 7 days and automatically cleaned up during sync/uninstall operations.
- Add internal/trash package (MoveToTrash, List, Cleanup, Restore, FindByName)
- Add `skillshare trash list` and `skillshare trash restore <name>` commands
- Show reinstall hint from metadata when uninstalling remote skills
- Add trash status to `skillshare doctor` output
- Opportunistic cleanup of expired items during sync and uninstall
- Support both global (~/.config/skillshare/trash/) and project (.skillshare/trash/) modes
- 10 unit tests + 10 integration tests
- Update docs: uninstall, trash, doctor, FAQ, file-structure, sync, daily-workflow
- new: redesign SKILL.md template based on Anthropic's skill-building
guide — description now includes WHAT+WHEN trigger format, step-based
instructions (Gather/Execute/Validate), examples, and troubleshooting
- search: prompt for keyword when run without arguments instead of
auto-browsing popular skills
- playground: default shell to $HOME (/sandbox-home) and mount source
as read-only to prevent accidental host file modifications