skill-manager
Manage, sync, and publish Agent Skills across multiple AI platforms (Claude, Codex, Gemini, Copilot) and marketplace repositories. Use when users want to list skills, sync between platforms, publish to marketplace keys, mirror canonical skills, audit drift, or set up their environment. Triggers on phrases like "list skills", "sync skills", "publish skill", "skill marketplace", "deploy skill", "audit skills", or "skill inventory".
What this skill does
# Skill Manager Manage repo-canonical Agent Skills across marketplace repositories and runtime install surfaces. ## Core Capabilities 1. **Runtime Install** - Install or refresh released skills into the canonical machine runtime path 2. **Development Sync** - Sync unpublished local skills across Claude Code, OpenAI Codex, Gemini CLI, and GitHub Copilot 3. **Inventory** - Discover and list all skills across platform locations 4. **Publish** - Push skills to tiered GitHub marketplace repositories (private/team/public) 5. **Audit** - Compare skill versions, detect drift, identify inconsistencies 6. **Validate** - Check skill compliance with the Agent Skills specification ## Source of Truth Skill-manager uses a **repo-canonical** model. The intended release flow is: 1. author in a git repo checkout 2. publish to a marketplace repo 3. install the published artifact into `~/.claude/skills/<skill>` 4. mirror `~/.codex/skills/<skill>` to the Claude install 5. if the skill materializes repo-local files, run its repo refresh flow separately Do not treat marketplace caches under `~/.claude/plugins/marketplaces/...` as the canonical runtime path. `scripts/sync.py` still exists for development-only movement of unpublished local skills. It is not the canonical release/install path. ## Supported Platforms | Platform | User Skills Path | Detection | |----------|-----------------|-----------| | Claude Code | `~/.claude/skills/` | `~/.claude/` exists | | OpenAI Codex | `~/.codex/skills/` | `~/.codex/` exists | | Gemini CLI | `~/.gemini/skills/` | `~/.gemini/` exists | | GitHub Copilot | `~/.copilot/skills/` | `~/.copilot/` exists | ## Quick Start ### First-Time Setup Run the interactive setup wizard: ```bash scripts/init.py ``` This will: 1. Detect installed AI platforms on your system 2. Configure which runtime platforms to manage 3. Set the canonical runtime policy (`claude` primary, optional Codex mirror) 4. Set development sync mode for ad hoc unpublished skill syncs 5. Optionally configure marketplace repositories ### Install Released Skills Into Runtime Install a released skill from a local marketplace clone into the canonical machine runtime path: ```bash scripts/install-from-marketplace.py --marketplace team --skill my-skill ``` This command: 1. reads the published artifact from the local marketplace clone 2. fully replaces `~/.claude/skills/<skill>` 3. recreates or verifies `~/.codex/skills/<skill>` as a symlink to the Claude install 4. removes stale files that disappeared from the published artifact Use `--all` to refresh every published skill from a marketplace clone. ### Sync Local Development Skills Between Platforms For unpublished or branch-only work, you can still sync a local skill directly: ```bash scripts/sync.py /path/to/local/my-skill --to codex ``` Or sync all skills from a specific runtime platform: ```bash scripts/sync.py --all --from-platform claude ``` Options: - `--to claude,codex,gemini,copilot` - Override target platforms - `--to auto` - Auto-detect installed platforms - `--to all` - Sync to all known platforms - `--mode symlink|copy` - Override development sync mode - `--from-platform claude|codex|gemini|copilot` - Required for `--all` unless legacy config still sets one - `--dry-run` - Preview changes without applying - `--force` - Overwrite existing skills without prompting - `--all` - Sync all skills from source platform ### List All Skills ```bash scripts/inventory.py ``` Options: - `--platform claude|codex|gemini|copilot|all` - Filter by platform - `--format table|json|yaml` - Output format - `--verbose` - Show full paths and metadata - `--source global|project|marketplace` - Include selected source(s), repeatable - `--exclude-source global|project|marketplace` - Exclude selected source(s), repeatable - `--global-only|--project-only|--marketplace-only` - Single-source shortcuts - `--no-marketplace` - Exclude marketplace entries - `--include-marketplace` - Backward-compatible alias (marketplace is already included by default) ### Inventory Behavior - Default `scripts/inventory.py` output includes all discovered sources: - global skills - project skills - marketplace skills - Use source flags only when you want to narrow results. - Legacy include forms remain supported: - `--include-marketplace` - `--include marketplace` - `--include=marketplace` ## Marketplace System `skill-manager` works with marketplace keys from `config.json` (for example: `private`, `team`, `public`, `partner`, `enterprise-core`). Keys are identifiers only; repository names and plugin names can be arbitrary. ### Publish a Skill ```bash scripts/publish.py ~/.claude/skills/my-skill --to team scripts/publish.py ~/.claude/skills/my-skill --to partner --in-development ``` Standard publish will: 1. Copy the skill to your team marketplace repo 2. Update marketplace.json 3. Validate and refresh the README.md "Available Skills" table entry 4. Push the publish branch to origin 5. Create and merge a PR back to the tracked default branch 6. Sync to other platforms (if configured) In-development publish (`--in-development`) will: 1. Copy the skill to `skills-in-development/` 2. Skip `marketplace.json` updates (so it is not installable from marketplace) 3. Skip automatic README table updates 4. Create a PR for review If the source path includes `skills-in-development/`, publish mode is inferred automatically. If you don't specify `--to`, you'll be prompted to choose. **README rule:** For standard publishes, the repository README.md must stay current. `publish.py` now validates the "Available Skills" table and updates the skill row automatically. If the README is missing or malformed, publish should fail until the repo documentation is fixed. ### Sync Marketplace Repos Pull latest from your marketplace repositories: ```bash scripts/marketplace-sync.py scripts/marketplace-sync.py --marketplace team scripts/marketplace-sync.py --status scripts/marketplace-sync.py --status --marketplace team scripts/marketplace-sync.py --branch master scripts/marketplace-sync.py --auto-stash scripts/marketplace-sync.py --prune-merged ``` `marketplace-sync.py --status` should be your first repo hygiene check. It reports: - Current checked-out branch versus the tracked default branch - Ahead/behind drift from `origin/<default-branch>` - Uncommitted changes - Extra local branches - Merged local branches that are safe cleanup candidates Use `--prune-merged` to delete local branches that are already merged into the tracked branch. Review unmerged extra branches manually before deleting them. ### Mirror Canonical Skills to Another Marketplace Mirror from a canonical source ref (`tag`, `branch`, or `commit`) into a target marketplace with manifest and README updates: ```bash scripts/marketplace-mirror.py mirror --from public --to partner --source-ref v1.2.0 scripts/marketplace-mirror.py mirror --from public --to partner --source-ref 8ab12cd --skill skill-manager scripts/marketplace-mirror.py mirror --from public --to partner --source-ref main --dry-run ``` Generate drift reports at any time: ```bash scripts/marketplace-mirror.py drift --from public --to partner scripts/marketplace-mirror.py drift --from public --to partner --source-ref v1.2.0 ``` ### Audit Runtime Drift Audit how a skill moves from repo source to marketplace artifact to runtime install: ```bash scripts/audit-runtime.py my-skill --marketplace team scripts/audit-runtime.py my-skill --marketplace team --source ~/GitHub/skills-team/skills/my-skill ``` This reports: - repo source vs marketplace artifact - marketplace artifact vs Claude runtime copy - Codex symlink status vs the Claude runtime copy ### Legacy Marketplace Distribution `marketplace-distribute.py` remains available for advanced or legacy workflows that intentionally symlink marketplace clones into other platform skill directories. It is not the recommended release/install path for Claude plus Codex. ### Audi
Related in Cloud & DevOps
appbuilder-action-scaffolder
IncludedCreate, implement, deploy, and debug Adobe Runtime actions with consistent layout, validation, and error handling. Use this skill whenever the user needs to add actions to an App Builder project, understand action structure (params, response format, web/raw actions), configure actions in the manifest, use App Builder SDKs (State, Files, Events, database), deploy and invoke actions via CLI, debug action issues, or implement patterns such as webhook receivers, custom event providers, journaling consumers, large payload redirects, action sequence pipelines, and Asset Compute workers. Also trigger when users mention serverless functions in Adobe context, action logging, IMS authentication for actions, or cron-style scheduled actions.
orchestrating-datacloud
IncludedSalesforce Data Cloud product orchestrator for connect→prepare→harmonize→segment→act workflows. Use this skill when the user needs a multi-step Data Cloud pipeline, cross-phase troubleshooting, or data space and data kit management. TRIGGER when: user needs a multi-step Data Cloud pipeline, asks to set up or troubleshoot Data Cloud across phases, manages data spaces or data kits, or wants a cross-phase sf data360 workflow. DO NOT TRIGGER when: work is isolated to a single phase (use the matching phase-specific skill), the task is STDM/session tracing/parquet telemetry (use observing-agentforce), standard CRM SOQL (use querying-soql), or Apex implementation (use generating-apex).
github-project-automation
IncludedAutomate GitHub repository setup with CI/CD workflows, issue templates, Dependabot, and CodeQL security scanning. Includes 12 production-tested workflows and prevents 18 errors: YAML syntax, action pinning, and configuration. Use when: setting up GitHub Actions CI/CD, creating issue/PR templates, enabling Dependabot or CodeQL scanning, deploying to Cloudflare Workers, implementing matrix testing, or troubleshooting YAML indentation, action version pinning, secrets syntax, runner versions, or CodeQL configuration. Keywords: github actions, github workflow, ci/cd, issue templates, pull request templates, dependabot, codeql, security scanning, yaml syntax, github automation, repository setup, workflow templates, github actions matrix, secrets management, branch protection, codeowners, github projects, continuous integration, continuous deployment, workflow syntax error, action version pinning, runner version, github context, yaml indentation error
sf-datacloud
IncludedSalesforce Data Cloud product orchestrator for connect→prepare→harmonize→segment→act workflows. TRIGGER when: user needs a multi-step Data Cloud pipeline, asks to set up or troubleshoot Data Cloud across phases, manages data spaces or data kits, or wants a cross-phase `sf data360` workflow. DO NOT TRIGGER when: work is isolated to a single phase (use the matching sf-datacloud-* skill), the task is STDM/session tracing/parquet telemetry (use sf-ai-agentforce-observability), standard CRM SOQL (use sf-soql), or Apex implementation (use sf-apex).
fabric-cli
IncludedUse this skill for Fabric.so CLI workflows with the `fabric` terminal command: diagnose/install/login, search or browse a Fabric library, save notes/links/files, create folders, ask the Fabric AI assistant, manage tasks/workspaces, generate shell completion, check subscription usage, produce JSON output, and use Fabric as persistent agent memory. Do not use for Microsoft Fabric/Azure/Power BI `fab`, Daniel Miessler's Fabric framework, Python Fabric SSH, Fabric.js, or textile/fashion fabric.
lark
IncludedLark/Feishu CLI skills: lark-cli operations for docs, markdown, sheets, base, calendar, im, mail, task, okr, drive, wiki, slides, whiteboard, apps, approval, attendance, contact, vc, minutes, event. Use when the user needs to operate Lark/Feishu resources via lark-cli, send messages, manage documents, spreadsheets, calendars, tasks, OKRs, deploy web pages, or any Feishu/Lark workspace operations.