Claude Skills Library MCP Server by frankxai

<p align="center"><img src=".github/assets/hero.svg" width="100%" alt="Claude Skills Library hero banner"/></p> <div align="center">

Claude Skills Library

116 Agent Skills for every skills-aware runtime, and 8 enforcement packs for Claude Code

Skills teach an agent a domain. Packs make the agent go through the gate. This is the only public library that ships both, with provenance pinned to commits, counts that check themselves in CI, and hooks that block "done" until the evidence exists.

Install · Packs · Catalog · Web hub · Strategy · Contribute · FAQ

Subscribe to the Waitlist

</div>

---

[!TIP] Free before you pay. Six free assets sit beside the paid packs and none is a teaser: a worked revenue loop with its arithmetic and a 30-day kill criterion, a real claims audit against a live page with file and line numbers, three prompt patterns that survive regeneration, one complete business pack, and the MIT open-core plugin kernel. Start with one.

Where to start

Two doors are open right now:

• Operator lab waitlist: join the waitlist at frankx.ai/community.
• Blueprint request access: request access to the Blueprint.

For the skills themselves, see the skill ranking for which ones are best formed today.

Install

[!IMPORTANT] Skills that ship with their evals. The first skill from this library is being packaged with its full promptfoo eval suite and the run output that shows exactly where it fails. If you would pay for skills that arrive proven, say which one should be first and what that is worth to you: join the eval-pack list. Two steps, every question skippable. Founding members name the next skill to be evaluated.

Three ways in. Pick one.

# 1. Claude Code plugin — auto-updating, skills discovered on their own
/plugin marketplace add frankxai/claude-skills-library
/plugin install claude-skills-library@claude-skills-library

# 2. skills.sh — the top-level skills, or a curated lane
#    (the skills nested under anthropic/, creative/ and technical/ come via the plugin or a clone)
npx skills add frankxai/claude-skills-library
npx skills add frankxai/skills            # AI architects: MCP, orchestration, model routing
npx skills add frankxai/creator-skills    # creators: video, music, images, brand voice

# 3. Clone and copy only what you want
git clone https://github.com/frankxai/claude-skills-library.git ~/claude-skills-library
mkdir -p ~/.claude/skills
cp -r ~/claude-skills-library/free-skills/mcp-architecture ~/.claude/skills/   # one
cp -r ~/claude-skills-library/free-skills/* ~/.claude/skills/                  # all

Some skills are namespaced (free-skills/anthropic/pdf). Copy the folder that directly contains the SKILL.md. Restart Claude Code and the skills are discovered.

Then work. Skills fire on their own from their description, or you name one:

"Design an MCP server with the mcp-architecture skill."
"Review this PR with the github-code-review skill."
"Run the verification-quality skill before you call this done."

---

What is here

Two layers. The first is what every skills repo has. The second is what none of the large ones ship.

Layer	What it is	Where
Skills	116 self-contained SKILL.md folders. Spec-validated frontmatter, trigger-rich descriptions, deep guidance in references/ loaded only on use.	free-skills/ · catalog
Packs	8 installable contracts. Each bundles the skills for one kind of work with whatever enforcement it has: hooks, CI scripts, drift checks, or templates. The packs table below says which.	packs/

A skill is loaded when the task matches its description. That makes it available, not mandatory. A pack adds what turns a suggestion into a default, and the mechanism differs by pack. web-excellence installs three hooks: a SessionStart note, a PreToolUse reminder on the first UI edit, and a Stop hook that blocks once if UI changed with no audit, each shipped with the test that proves it cannot be satisfied by its own reminder text. estate-guard installs a PreToolUse gate that denies the hard stops and a taint hook for fetched content. routine-contract ships a CI workflow and a digest primitive that fails the run when nothing durable was written, for each routine you wire to it. risk-classifier ships a fail-closed classifier script for you to wire into CI. agent-infrastructure installs skills and checks installed copies against canon. film-excellence installs a skill and templates and is advisory.

---

Packs: skills that enforce

Installed into 19 repositories across the FrankX, Starlight, and Arcanea estate as of 2026-10-09. Re-run an installer to upgrade. web-excellence, routine-contract and agent-infrastructure set an installed skill that differs aside under .claude/skills/.replaced/<name>.<timestamp>/ and never overwrite an earlier set-aside; agent-hygiene sets it aside under .claude/skills/.replaced/<name>/ and keeps only the most recent copy; estate-guard and film-excellence overwrite their skill in place. Commit a customised copy before re-running any of the last three. All but risk-classifier accept --dry-run to show the diff first.

Pack	What it gates	Enforced by
web-excellence	Website and web-design work. Sequences a live Vercel Web Interface Guidelines audit, motion review, Core Web Vitals, and before/after screenshots at 375 / 768 / 1440. "Done" needs evidence, never a self-assigned score.	11 vendored skills pinned to upstream commits (Vercel, Emil Kowalski, nextlevelbuilder, arvindrk, Leonxlnx), 3 native skills, 3 hooks, 18 tests
agent-hygiene	The agentic surface's own quality: agent definitions, skills, commands, AGENTS.md, CLAUDE.md. One skill states the bar (the estate guardrails, the agent-file contract, model and effort routing); a linter ratchets it on every PR that touches such a file; a Stop hook requires each hand-back after a change to end with a Verified: line naming a command that actually ran.	Skill, 17-check linter, PR workflow, 1 hook, 90 tests
estate-guard	The agentic surface: workflows, hooks, settings, MCP configs, skills, API routes. Denies force-push to main, recursive deletes, curl | sh, permission bypass. Marks instruction-shaped text in fetched or MCP output as data.	Scanner, 3 hooks, weekly CI, severity-triaged fix patterns
agent-infrastructure	The multi-agent operational layer: swarm orchestration, hive-mind, ReasoningBank memory, hooks, stream chaining, verification, skill tooling. Nine skills that drifted into 2 to 4 divergent copies across five repos now have one canonical source.	Manifest plus installer over free-skills/, content hashes, drift check
risk-classifier	PR readiness. Classifies every changed file against eight risk classes. Unknown fails closed. Secret-shaped content anywhere fails closed.	CI script, classes.json, tests
routine-contract	Scheduled agents. Once a routine is wired to the digest primitive, it cannot report SUCCEEDED while committing nothing: preflight proves the output sink exists, the primitive exits non-zero if it could not persist, and the drift detector catches a routine that stopped reporting. The installer puts the pieces in place; you enumerate the routines and point each at the primitive.	Skill, digest script, fleet drift detector, daily CI, 12 tests
film-excellence	Narrative film and short-form video. Doctrine, dialogue tells that mark machine writing, token contract, crew chairs, eight-stage pipeline. The film counterpart to web-excellence.	Native skill, templates, design and taste contracts

git clone --depth 1 https://github.com/frankxai/claude-skills-library /tmp/csl
/tmp/csl/packs/web-excellence/install.sh "$PWD"
/tmp/csl/packs/estate-guard/install.sh "$PWD"

Then one line in the repo's CLAUDE.md: Website / web-design work goes through the web-release-gate skill first.

---

Why this library

The largest skills repositories are excellent at judgment: how to think about a design, a diff, a test. They are weak at proof. None of them measures anything, produces evidence at real breakpoints, refuses to let work be called done, or tells you where a vendored skill came from and whether it has moved upstream since. This library is built around those gaps.

	Most skills repos	This library
Skill format	SKILL.md	SKILL.md, validated in CI for frontmatter, name, description length, filename
Provenance	Copied	Third-party skills in the packs are pinned to an upstream commit and re-syncable, with the upstream license beside each of the eleven (web-design-guidelines carries the upstream's README licence declaration and commit, since that repository ships no LICENSE file). The Anthropic reference skills under free-skills/anthropic/ carry their licenses but not yet pins; closing that is bet 3 in the strategy
Enforcement	Advisory	Hooks that block once without an audit, each with a test
Counts	Hand-maintained	The README's skill total, pack total, and per-category counts are checked against the tree; a mismatch fails the build
Inbound work	Paste it in	A four-gate absorption contract (license, provenance, distinctness, attestation) that refuses rather than warns
Scheduled agents	"Report your findings"	A digest primitive that fails the run if nothing durable was written

The honest gap on our side: evals. All 116 in this library read well. None yet ships evidence that it still behaves under pressure. That is the next moat and the first bet in docs/STRATEGY.md.

---

Skill categories

Counts are the groupings in docs/CATALOG.md, regenerated from the tree by scripts/generate_catalog.py. 116 skills across 11 categories.

Category	Count	Examples
AI agents and orchestration	17	agent-hygiene, agentic-orchestration, swarm-orchestration, model-routing, v-swarm
AI frameworks, MCP and SDKs	13	mcp-architecture, mcp-builder, openai-agentkit, claude-sdk, langgraph-patterns
Oracle and cloud	8	oracle-ai-architect, oracle-database-expert, oci-services-expert, ai-architecture
Web, frontend and animation	14	nextjs-expert, ui-ux-design-expert, tailwind, gsap, three, framer-expert
Engineering workflow and GitHub	10	github-code-review, performance-analysis, verification-quality, hooks-automation
Content, writing and brand	14	brand-voice, book-publishing, social-media-strategy, creator-productivity
Creative and media production	22	suno-ai-mastery, video-production-workflow, hyperframes-media, higgsfield-soul-id
Mind, body and philosophy	5	greek-philosopher, spartan-warrior, gym-training-expert, health-nutrition-expert
Documents and productivity	6	pdf, docx, pptx, xlsx, product-management-expert, webapp-testing
Meta and library	1	contribute-catalog
Other	6	feynman-thinking, loop-designer, loop-orchestrator, loop-runner

Browse the full catalog, or the generated site at docs/index.html.

---

A few to start with

Skill	Why you would reach for it
mcp-architecture	Design an MCP server's resources, tools, prompts, and security from first principles
ai-architect-review	Review any system that calls a language model: agent loops, retrieval paths, tool surfaces
nextjs-expert	App Router, server/client boundaries, caching, and the production gotchas that bite
github-code-review	Turn a raw diff into a correctness, security, and style review with actionable comments
model-routing	Route a task to the cheapest model that can do it well, with the rules written down
oracle-database-expert	Oracle 23ai, Autonomous DB, AI Vector Search, SQL and PL/SQL tuning, HA
suno-ai-mastery	Prompt-engineer commercial-quality music with Suno v4.5+
loop-designer	Compile a goal into a file-backed agent loop any coding agent can run unattended
greek-philosopher	Socratic questioning and a Stoic read on a hard decision

---

How the pieces fit

flowchart LR
    SIS["Starlight Intelligence System<br/>substrate: memory, agents, SIP"] --> ACOS["Agentic Creator OS<br/>canonical skill source"]
    ACOS -->|"mirror the strongest"| LIB["claude-skills-library<br/>116 skills · 8 packs"]
    UP["Upstreams<br/>Vercel · Emil Kowalski · Anthropic · ..."] -->|"pinned, licensed, drift-checked"| LIB
    LIB -->|"/plugin · npx skills add · cp"| CC["Claude Code"]
    LIB -->|"runtimes/"| RT["Cursor · Codex · Gemini CLI<br/>OpenCode · Antigravity"]
    LIB -->|"packs/*/install.sh"| REPOS["19 estate repos<br/>hooks + CI + tests"]

Starlight is the substrate. ACOS is where skills are authored and battle-tested first. This repo is the public, install-from catalog, and the packs are how its standards reach every other repo.

---

Skills vs. MCP vs. prompts

	A prompt	An MCP server	An Agent Skill
What it is	Text you paste	A connection to tools and data	A model-invoked capability
Gives the agent	Instructions, once	New actions (APIs, files, DBs)	Expert knowledge plus workflow
Activation	Manual, every time	Always on once connected	Auto-loads when relevant
Token cost	Every turn	Tool schemas always present	Near zero until triggered
Portability	One chat	A server you run	A versioned folder, shared across runtimes

Use a skill to teach the agent how to do something well. Use MCP to give it access to a system. Use a prompt for a one-off. Many skills here pair with an MCP server.

---

Works with six runtimes

Agent Skills are a portable format. One adapter guide per runtime lives in runtimes/.

Runtime	Maturity	Guide
Claude Code	Native	runtimes/claude-code.md
Antigravity (agy)	Substrate-compatible	runtimes/antigravity.md
OpenCode	Adapter pattern	runtimes/opencode.md
Codex CLI	Per-skill MCP wrap	runtimes/codex.md
Gemini CLI	Per-skill template	runtimes/gemini-cli.md
Cursor	Per-skill convert	runtimes/cursor.md

The rationale for MCP as the universal bridge is in MULTI_RUNTIME.md.

---

Anatomy of a skill

free-skills/<skill-name>/
├── SKILL.md          # required: frontmatter + instructions
├── references/       # optional: deep docs loaded on demand
├── scripts/          # optional: executable helpers the agent runs
└── assets/           # optional: templates, fonts, icons used in output

---
name: mcp-architecture          # lowercase, hyphenated, 64 chars or fewer
description: Design and implement MCP servers... Use when architecting an MCP server...
---

The description is the most important line. It is what the model uses to decide when to load the skill. State what it does and when to use it, with trigger keywords.

Authoring standard: spec/README.md. Start here: template/SKILL.md.

---

Quality and validation

Zero-dependency scripts enforce the standard on every push and pull request (validate.yml, pack-tests.yml, catalog-check.yml):

python3 scripts/validate_skills.py           # frontmatter + structure, exits non-zero on failure
python3 scripts/generate_catalog.py --check  # docs/CATALOG.md matches the tree
python3 scripts/generate_site.py --check     # docs/index.html matches the tree
python3 scripts/check_counts.py              # README counts match the tree
python3 scripts/check_links.py               # no broken internal links
node scripts/build-catalog.mjs --check       # catalog/skills.json matches the marketplace
node --test scripts/tests/absorb.test.mjs    # the four absorption gates still refuse
node --test scripts/tests/harness-e2e.test.mjs # harness end-to-end tests

Inbound external work runs through ABSORPTION.md: license, provenance, distinctness, attestation. node scripts/absorb.mjs refuses rather than warns. What has entered is recorded in ABSORBED.md.

[!NOTE] Pinned upstream, 2026-10-10. affaan-m/ECC at 4eb71d92a39cab44ad40ac9d8a6a5ccb4029d6c2 (MIT) is a dependency of the Claude config repo, not a prose dump in free-skills/. Patterns that cleared the four gates are listed in ABSORBED.md. The port receipt is docs/ecc-port-report.md. Watching a repo is not the same as absorbing it.

---

Catalog site

scripts/generate_site.py renders the catalog to docs/index.html. It deploys as a static site to Vercel with the committed vercel.json, or to GitHub Pages from the docs/ folder.

---

Strategy

The plan to make this the reference skills library, in one page: docs/STRATEGY.md. The short version:

1. Evals before volume. Ship the eval suite, the passing cases, and the failing ones, for the top skills first.
2. Packs are the product. Enforcement, provenance, and tests are what a skills directory cannot give you.
3. Absorb patterns, not prose. The best skills from other libraries enter through the four gates, with credit, or are depended on, never pasted.
4. One canonical copy. Skills that exist in several estate repos get mirrored here and installed back from here.

---

Community

• Ask a question, share a working example, or suggest an idea.
• Report a bug or request a skill with the affected skill and a small, shareable example.
• Contribute a fix or skill. Evidence from real use and clear source attribution are what get a PR merged.
• Report a vulnerability privately.

See SUPPORT.md for routing and CODE_OF_CONDUCT.md for participation. Remove credentials and private material before posting.

---

Contributing

1. Read CONTRIBUTING.md and the authoring standard.
2. Copy template/ to free-skills/<skill-name>/ and write your SKILL.md.
3. Run python3 scripts/validate_skills.py and python3 scripts/generate_catalog.py until both are clean.
4. Open a pull request.

---

Related repositories and web guides

• FrankX Skills Hub: interactive Skill Builder for generating SKILL.md files and curated domain skill overview.
• Claude Code Skills: The 10 You Actually Need and AI Architecture Blueprints: field notes and reference topologies from production deployments.
• frankxai/skills and frankxai/creator-skills: the curated, actively maintained lanes for AI architects and for creators. This library is the full catalog behind them.
• frankxai/agentic-creator-os: the canonical, larger source for skills, agents, and commands. New skills are authored and battle-tested there first.
• frankxai/Starlight-Intelligence-System: the substrate. Persistent context, memory, and the multi-agent layer the skills run on.
• frankxai/starlight-design-intelligence: the design and editorial kernel the web-excellence pack enforces.

---

FAQ

Does loading 116 skills bloat my context? No. Only each skill's name and description is preloaded for routing. The body loads when a skill triggers, and references/ load only when read. A library of 116 costs almost nothing until a skill fires.

How does the right skill get picked from a library this large? By the description. Every skill states what it does and when to use it with explicit trigger keywords. If one mis-triggers, sharpen its description. PRs welcome.

Do I have to install all of them? No. Install the plugin for everything, npx skills add a curated lane, or copy the folders you want.

Is this affiliated with Anthropic? No. It is an independent, MIT-licensed library built on Anthropic's open Agent Skills format. Skills under free-skills/anthropic/ are upstream reference skills that retain their original licenses.

Can I use these outside Claude Code? Yes. See runtimes/ for Antigravity, OpenCode, Codex, Gemini CLI, and Cursor.

---

Would you pay for a skill that proves it works?

Every first-party skill here is free and MIT. That is not changing. Skills under free-skills/anthropic/ keep Anthropic's own terms.

A skill is prose until something tests it. The idea on the table: a pack where every skill ships with its eval suite, the cases it passes, and the cases it fails. There is no paid product today. Before building one, say whether it should exist, and name the band.

Tell us what it is worth

The first fifty people to answer get one skill with its full eval suite and run output, free. At fifty answers the eval harness is published open-source regardless.

---

License

MIT. See LICENSE. Imported skills retain their upstream licenses, recorded in THIRD_PARTY_NOTICES.md and in each vendored folder.

---

<div align="center"> <img src="assets/logo.svg" width="64" alt="Claude Skills Library mark"/>

Join the eval-pack list · Web hub · Star the repo · Catalog · Issues · Discussions

</div>