About Llms
Llms is an MCP server published by matteocervelli in the Developer Tools category: centralized LLM configuration and documentation management system. Tools for building skills, commands, agents, prompts, and managing MCP servers. Multi-LLM support (Claude Code, Codex, OpenCode).
The repository has 25 stars and 2 forks, with the last commit 5 months ago. Recent commit activity is the strongest signal that reported issues will get attention.
Install
git clone https://github.com/matteocervelli/llmsThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Ask AI
Ask AI about Llms
Powered by Claude · Grounded in docs
Security checks
Static checks derived from the repository 5 days ago. Heuristics on the source, not an audit.
- ✓README presentThe repository ships documentation.
- ✓License declaredUsable under a stated license.
- ✓Tests presentA test suite exists in the repository.
- !Dependencies pinnedNo lockfile. Installs can pull different code over time.
- ✓No dynamic code executionSource scan found no eval, dynamic Function, or shell execution.
- !Scoped permissionsDoesn't declare a permission scope. Assume it can do anything its process can.
Releases
README
llms
Production-ready Claude Code and Codex configuration — skills, hooks, rules, and agents used in real AI-driven development workflows.
What's inside
| Directory | Contents |
|---|---|
.claude/ |
Claude Code skills, hooks, rules, agents, settings |
.codex/ |
Codex equivalent — same skills and hooks, Codex config |
docs/anthropic/ |
Fetched Anthropic reference docs (API, Claude Code, MCP) |
.archive/ |
Previous versions — builder tools v0.1, Feature-Implementer v2 |
Quick start
git clone https://github.com/matteocervelli/llms.git
cd llms
# Copy to your project
cp -r .claude/ /your-project/.claude/
# Or install globally
cp -r .claude/skills/* ~/.claude/skills/
cp -r .claude/rules/* ~/.claude/rules/
cp -r .claude/hooks/* ~/.claude/hooks/
Skills (39)
Full catalog with trigger phrases: docs/skill-catalog.md
SDLC Pipeline
/discovery /story /story-verify /spec /design /implementation /fix /quick /diagnose
Quality & Security
/code-review /pre-commit /quality-check /security /security-verify /supply-chain-audit /docker-audit /test-scaffold
Git / PR / Release
/ship /pr-creator /pr-fix /pr-merge /release /deploy /progress
Docs & Ops
/docs /documentation-updater /deps /health /ops /techdebt /website-health
Meta
/map-codebase /skillify /registry /review /project-create /analytics /frontend /claude-code-guide
Hooks
Unified dispatcher — one entry point, specialized handlers.
.claude/hooks/
hook_handler.py # dispatcher
handlers/
bash.py # block dangerous commands
file.py # protect sensitive files
context_monitor.py # warn at 65/75% context usage
plan.py # enforce TaskCreate after plan approval
rules/
blocked_commands.yaml
protected_files.yaml
formatters.yaml
Wired in .claude/settings.json.
Rules (21)
tdd safety security-gate code-quality dependencies defense-in-depth anti-slop docker python-tools scoped-testing temporal-awareness skill-architecture claude-md-structure claude-md-branching lockfile-safety package-freshness frontend-verification release-workflow github-workflow github-cli migration-safety
Codex
.codex/ mirrors .claude/ with Codex-specific config:
codex --config .codex/config.toml
See .codex/instructions.md for Claude Code vs Codex differences.
Versions
| Version | Description |
|---|---|
| v0.2.0 | Production skill library — 39 skills, unified hook dispatcher, Codex mirror |
| v0.1.x | Feature-Implementer v2 (archived in .archive/) |
Support
License
README fetched from the repository 5 days ago. The original is authoritative.