A collection of skills for AI coding agents. Skills are packaged instructions and scripts that extend agent capabilities.
This repository also configures its own agents. See what agent-skills tells them →
Skill Claude CodeCodex
Systematic JavaScript/TypeScript performance audit and optimization using V8 profiling and runtime patterns. Use when (1) Users say 'optimize performance', 'audit performance', 'this is slow', 'reduce allocations', 'improve speed', 'check performance', (2) Analyzing code for performance anti-patterns (O(n²)…
Skill Claude CodeCodex
Comprehensive vitest testing guidance for TypeScript projects. Use when (1) Writing new tests with AAA pattern, parameterized tests, or async/await, (2) Reviewing test code for anti-patterns like loose assertions (toBeTruthy), over-mocking, or nested describe blocks, (3) Optimizing slow test suites, (4) Implementing…
Skill Claude CodeCodex
Extract explicit and implicit constraints (compliance, security, hosting, tooling, stakeholder, scope, and external-dependency boundaries) from a project's documentation, then build or update a canonical CONSTRAINTS.md. Spawns one subagent per source document to flag externally-imposed boundaries with evidence…
Skill Claude CodeCodex
Build or update a canonical EPISTEMIC-MAP.md sorting a project's knowledge into four states: validated Facts, open Questions, undocumented Assumptions, and inferred Risks. Purpose-built for prototype-to-production handoff, where a signed contract means a new engineering team must take over before the original…
Skill Claude CodeCodex
Generate fumadocs MDX documentation for agent skills. Use when users say "generate docs", "document this skill", "create fumadocs for X", "update skill docs", "validate docs", or when working with docs/content/docs directory. Handles initial generation, smart updates preserving manual edits, and validation. Uses SHA…
Skill Claude CodeCodex
Extract internal terminology, acronyms, shorthand, and jargon from a set of documents, then build or update a project glossary in JARGON.md. Reuse an existing JARGON.md anywhere in the repo if one already exists; otherwise default to the repo root. Spawns one subagent per input file to flag terms with plain English…
Skill Claude CodeCodex
Onboard a repository to agent-driven development by creating or refreshing a complete AGENTS.md or CLAUDE.md through behavior-focused discovery, structured interviewing, drift-aware updates, conflict-aware synthesis, proportional updates, and preview-before-write review. Use when the user wants to create, replace…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: