skill-fetch
This skill should be used when the user asks to "fetch skill", "install skill", "search for a skill", or when a hook outputs "MISSING EXTERNAL SKILL". Searches 9 registries (SkillsMP, GitHub, Anthropic Skills, ClawSkillHub, skills.sh, PolySkill, SkillHub, Skills Directory) with multi-variant search, quality scoring, security labels, pagination, and local/global installation.
How do I install this agent skill?
npx skills add https://github.com/girofu/skill-fetch --skill skill-fetchIs this agent skill safe to install?
- Gen Agent Trust Hubwarn
The skill is designed to search for, download, and install other AI agent skills from multiple online registries and GitHub repositories. While it incorporates a built-in security scanner to detect malicious patterns in downloaded content and requires user confirmation before installation, it performs several high-risk operations including the execution of remote packages via `npx`, the downloading of arbitrary code bundles, and the management of API keys stored in a local configuration file. Users should exercise caution and review the sources of skills being installed.
- Socketwarn
2 alerts: gptSecurity, gptAnomaly
- Snykwarn
Risk: MEDIUM · 2 issues
What does this agent skill do?
Skill Fetch
Search, score, and install agent skills from multiple registries in parallel.
When to Use
- A skill-eval hook outputs "MISSING EXTERNAL SKILL"
- The current task requires domain expertise not available locally
- The user asks to "fetch skill", "search for a skill", or "install a skill"
Critical Rules
- Never use
skillsmp_get_skill_contentto preview before deciding — search descriptions are sufficient. Only use as a fallback if installation fails. - Stop on first results, continue only on zero (max 5 rounds) — any round with ≥1 result proceeds to analysis.
- Only the user can decide to skip — the LLM must never skip installation on its own.
- Always use plain-text interaction — do not use AskUserQuestion. Output analysis and wait for the user to reply with a number or command.
- Show up to 5 results per page — with full analysis (content, pros, cons, repo URL) so the user can make an informed decision.
Platform Compatibility
This skill works across 6+ AI coding agents (Claude Code, Cursor, Codex, Gemini CLI, Windsurf, Amp). Tool names adapt automatically. See references/platform-adapters.md for the full mapping table.
Procedure
Step 0: SkillsMP MCP Self-Check (Claude Code only)
Before searching, verify that the SkillsMP MCP server is available:
- First, try loading deferred tools: Run
ToolSearch("skillsmp")to load any deferred SkillsMP MCP tools. In Claude Code, MCP tools are often deferred (not loaded until requested) and have namespaced names likemcp__skillsmp__skillsmp_search. TheToolSearchcall resolves these. - After ToolSearch, check if any
skillsmp_*ormcp__skillsmp__*tool is now available - If available → proceed to Step 1
- If still not available → run:
claude mcp add --scope user skillsmp -- npx -y skillsmp-mcp-server - Inform the user: "SkillsMP MCP server has been registered. It will be available after restarting the session. Continuing search with the remaining 7 sources for now."
- Proceed to Step 1 (SkillsMP sources will be skipped this session, but available in future sessions)
Tool name note: SkillsMP tools may appear with MCP namespace prefix:
mcp__skillsmp__skillsmp_searchinstead ofskillsmp_search. Both forms work — use whichever is available.
Non-Claude Code agents: Skip this step. SkillsMP tools are Claude Code-specific.
Step 0.5: Load API Keys (MANDATORY — execute before Step 2)
Sources 8-9 require API keys. Read the config file to determine availability:
- Read
~/.claude/skills/.fetch-config.jsonNOW — do not assume keys are absent without reading the file - Expected format:
{ "SKILLHUB_API_KEY": "sk-sh-...", "SKILLS_DIRECTORY_API_KEY": "sk_live_..." } - Record the result — set two flags for use in Step 2b:
HAS_SKILLHUB_KEY: true/falseHAS_SKILLS_DIR_KEY: true/false
- If config file does not exist → both flags = false, inform user they can create the file
- If config file exists but a specific key is missing → that flag = false
⚠️ COMMON FAILURE MODE: LLM skips reading the config file and assumes "no key". This causes Sources 8-9 to be incorrectly skipped even when keys ARE configured. Always read the file first.
Step 1: Determine Search Keywords and Source
URL mode: If $ARGUMENTS is a URL (starts with https://github.com/... or any https://), skip directly to Step 3 installation flow.
Search mode:
- Has
$ARGUMENTS: Use directly as search terms - No search terms (auto-triggered): Prefer
Suggested search termsfrom hook output, otherwise extract 2-3 queries from task context
Step 2: Parallel Search — ALL 9 Sources (mandatory)
⚠️ MANDATORY: You MUST fire ALL 9 sources. Do NOT proceed to scoring until all 9 sources have returned or failed.
⚠️ COMMON FAILURE MODE: LLM fires sources 1-2 (SkillsMP), gets results, then skips sources 3-9. This is WRONG. SkillsMP results alone are insufficient — GitHub, ClawhHub, skills.sh, and PolySkill contain different skills not indexed by SkillsMP.
⚠️ EXECUTION ORDER: You MUST complete BOTH Step 2a AND Step 2b before proceeding to Step 2.5. Step 2a alone is NOT sufficient.
Step 2a: Fire SkillsMP sources (1-2)
| # | Source | Tool Call | Fallback |
|---|---|---|---|
| 1 | SkillsMP AI | skillsmp_ai_search × 3 query variants (parallel) | Skip if MCP unavailable |
| 2 | SkillsMP keyword | skillsmp_search(query) | Skip if MCP unavailable |
Step 2b: Fire non-SkillsMP sources (3-9) — DO NOT SKIP
⚠️ You MUST fire these sources even if Step 2a already returned results. These sources contain skills NOT in SkillsMP.
| # | Source | Tool Call | Fallback |
|---|---|---|---|
| 3 | GitHub repos | gh search repos "{query}" --json name,description,url,stargazersCount,updatedAt --limit 5 --sort stars (do NOT append "skill SKILL.md") | gh search code "{query}" --filename SKILL.md --limit 5 |
| 4 | Anthropic Skills | gh search code "{query}" --repo anthropics/skills --filename SKILL.md --limit 5 | gh api tree fallback |
| 5 | ClawhHub | npx -y clawhub search "{query}" | Skip on failure |
| 6 | skills.sh | WebFetch("https://skills.sh/api/search?q={query}&limit=5") | curl -s via Bash |
| 7 | PolySkill | npx -y @polyskill/cli search "{single_keyword}" --limit 5 (extract most specific single keyword from query — multi-word queries return 0) | Skip on failure (no REST API) |
| 8 | SkillHub | If HAS_SKILLHUB_KEY = true (from Step 0.5): bash {SKILL_BASE_DIR}/scripts/fetch-skillhub.sh "{query}". If false: npx -y @skill-hub/cli search "{query}" --limit 5 (timeout: 10000) | CLI fallback on failure |
| 9 | Skills Directory | If HAS_SKILLS_DIR_KEY = true (from Step 0.5): bash {SKILL_BASE_DIR}/scripts/fetch-skills-directory.sh "{query}". If false: skip. Never use curl directly or WebFetch. | Skip only if key is confirmed absent in Step 0.5 |
⚠️ Sources 8-9 REQUIRE Step 0.5 to have been executed. If Step 0.5 was skipped, go back and read
~/.claude/skills/.fetch-config.jsonNOW before marking Sources 8-9 as skipped.{SKILL_BASE_DIR}= the base directory of this skill (shown at the top when skill loads).
See
references/search-sources.mdfor detailed parameters, response formats, query variant examples, and curl fallback commands.
⚠️ POST-SEARCH CHECKLIST (mandatory before proceeding to Step 2.5):
Before scoring, output this checklist with actual status for EVERY source:
Step 0.5 config: HAS_SKILLHUB_KEY={true/false} HAS_SKILLS_DIR_KEY={true/false}
Sources queried: [1] SkillsMP AI ✅ [2] SkillsMP KW ✅ [3] GitHub ✅/❌ [4] Anthropic ✅/❌ [5] ClawhHub ✅/❌ [6] skills.sh ✅/❌ [7] PolySkill ✅/❌ [8] SkillHub ✅/❌ [9] Skills Dir ✅/❌
Validation rules — go back and fix before proceeding:
- If sources 3-7 are ALL ❌ → re-execute sources 3-7 NOW
- If
HAS_SKILLHUB_KEY=truebut [8] is ❌ → executebash {SKILL_BASE_DIR}/scripts/fetch-skillhub.shNOW - If
HAS_SKILLS_DIR_KEY=truebut [9] is ❌ → executebash {SKILL_BASE_DIR}/scripts/fetch-skills-directory.shNOW - If Step 0.5 was never executed (no config line above) → read
~/.claude/skills/.fetch-config.jsonNOW, then re-evaluate sources 8-9
After ALL sources return → deduplicate (see references/search-sources.md) → proceed to Step 2.5.
Round strategy (max 5 rounds): The "≥1 result → stop" rule applies to rounds, not individual sources. Within a single round, ALL 9 sources must be queried. Only if ALL 9 sources return 0 results in a round should you proceed to the next round with broader keywords. If any source returns ≥1 result in a round, proceed to scoring (do NOT start another round).
Step 2.5: Scoring and Ranking
Calculate a quality score (0-100) for each deduplicated result. See references/quality-signals.md for details.
Scoring formula: Total = Relevance(0-40) + Freshness(0-25) + Community(0-20) + Trust(0-15) + External Bonus(0-5)
Supplementary lookup: For the top 5 results, use gh api repos/{owner}/{repo} --jq '{pushed_at,stargazers_count}' to get update time and GitHub stars. Skip lookup for high-star (≥50) results with precisely matching descriptions. Maximum 3 gh api calls.
Grade labels: 🟢 85+ Strongly Recommended | 🟢 70-84 Recommended | 🟡 55-69 Worth Considering | 🟡 40-54 Marginal | 🔴 <40 Not Recommended
Safety Gate: Scan Before Installing (skills and MCP servers)
Anything from a third-party registry is untrusted until it has been scanned. An installed skill or MCP server runs with your agent's permissions, so scan before it lands in a place the agent loads from. Apply this gate to every install in Step 3 and to every MCP server this skill would register (including the SkillsMP MCP in Step 0).
- Stage first. Download into a fresh temp directory outside every skills directory (
mktemp -d), never directly into~/.claude/skills/,.claude/skills/or a plugin directory. Nothing in staging is loaded or run. - Scan the whole bundle. Every file, Categories A-F (see
references/interaction-patterns.md). Also check what loads or runs without a prompt: SKILL.md frontmatter (allowed-tools,hooks), bundled.mcp.jsonand hook configs, and anyscripts/the skill tells the agent to run. - Review an MCP server before registering it (
claude mcp add,.mcp.jsonor any config edit):- Source: the repository and package owner, last release, and whether the npm/PyPI package matches the repo it claims to come from.
- Version: register an exact version (
pkg@1.2.3), notnpx -y pkgor@latest. - Install scripts: read
postinstall/setup.pyor equivalent before anything executes. - Tools: list what each tool can do (read, write, delete, execute) and drop servers whose reach exceeds the task.
- Credentials and network: which keys it needs (use a narrowly scoped key, never a broader one) and which hosts it contacts.
- Show the result and wait. Report findings as file, category, severity (or "no findings"). A clean scan lowers
risk but does not prove safety. Critical findings (A, B) block the install; anything else needs the user's explicit
confirm install. The LLM must never confirm for the user. - Then install. Move the staged files to the location chosen in 3c (or register the pinned MCP server), record SHA-256 (3e), and delete the staging directory. If the bundle cannot be fetched or scanned in full, do not install.
Step 3: Analyze, Select, Install
See references/installation-guide.md for the complete installation workflow.
Quick reference — flow:
- 3a. Display sorted results (5 per page, score + pros/cons)
- 3b. Wait for user reply (number to install,
cfor next page,skipto end) - 3c. Ask installation location (G=global, L=local) — MANDATORY
- 3d. Execute installation through the Safety Gate above (stage → scan → install; SkillsMP and GitHub alike, GitHub → fetch the full bundle, not just SKILL.md, then scan every file before moving it into place). Use
scripts/fetch-skill-bundle.shfor GitHub URLs. - 3e. Post-install verification (file existence, frontmatter, conflict check, bundle completeness, SHA-256 on every installed file)
- 3f. Update
~/.claude/skills/.fetch-metadata.json - 3g. Activate and confirm with user
⚠️ Critical: When installing from GitHub, the skill is its entire directory —
SKILL.mdplusreferences/,scripts/,assets/,templates/,prompts/,data/, etc. Downloading onlySKILL.mdwill produce a broken install for any skill that delegates to those files. Always fetch the complete bundle.
Step 4: Digest the Installed Bundle
An installed skill is a directory, not a single SKILL.md. Many skills split their content across subdirectories — references/ (docs), scripts/ (helpers), assets/, templates/, prompts/, data/, examples/, etc. If you only read SKILL.md you may miss instructions the skill author expects you to load on demand.
- Use Glob to list the full installed tree:
{install-path}/{skill-name}/**/*. - Verify bundle completeness. If SKILL.md references relative paths (
bash scripts/foo.sh,See references/bar.md,Load assets/template.json) and those files are missing, the install is incomplete — re-run the GitHub bundle fetch (seereferences/installation-guide.md§3d). - Read files directly relevant to the current task (check the first 30 lines for relevance). Don't limit this to
references/— the relevant file may live underprompts/,templates/, or another subdir. - Summarize key knowledge for use in subsequent planning.
After completion, output: External skill installed successfully: {skill-name}
Completion Phrases
- Success:
External skill installed successfully: {name} - Skipped:
External skill fetch: user chose to skip installation.
Additional Resources
Rationalization Table and Red Flags are in
references/interaction-patterns.md. Consult when rationalizing skipping steps.
references/interaction-patterns.md— Output templates, user reply handling, security reviewreferences/quality-signals.md— Quality assessment dimensions, lookup methods, ranking algorithmreferences/search-sources.md— Source-specific commands, error handling, deduplication rulesreferences/platform-adapters.md— Cross-platform tool mapping, installation paths, fallback strategiesreferences/installation-guide.md— Complete Step 3 installation workflow (3a-3g)references/local-index.md— Local skill/plugin scan implementation for pre-search deduplicationscripts/fetch-skillhub.sh— SkillHub API search (reads key from~/.claude/skills/.fetch-config.json)scripts/fetch-skills-directory.sh— Skills Directory API search (reads key from config)scripts/fetch-skill-bundle.sh— Download a complete skill bundle (SKILL.md + all sibling files/subdirs) from any GitHubblob/tree/rawURL, preserving directory layout
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/girofu/skill-fetch/skill-fetch">View skill-fetch on skillZs</a>