agent-skills and ai-agent-skills

These are **complements**: A provides a general-purpose, validated skill registry for multiple AI coding agents, while B supplies pre-configured, project-specific skills that would typically be sourced from or integrated alongside registries like A.

agent-skills
75
Verified
ai-agent-skills
44
Emerging
Maintenance 25/25
Adoption 10/25
Maturity 20/25
Community 20/25
Maintenance 10/25
Adoption 10/25
Maturity 11/25
Community 13/25
Stars: 1,745
Forks: 174
Downloads:
Commits (30d): 145
Language: TypeScript
License:
Stars: 140
Forks: 15
Downloads:
Commits (30d): 0
Language: JavaScript
License: MIT
No risk flags
No Package No Dependents

About agent-skills

tech-leads-club/agent-skills

The secure, validated skill registry for professional AI coding agents. Extend Antigravity, Claude Code, Cursor, Copilot and more with absolute confidence.

Provides curated skills as structured markdown with templates and references rather than compiled code; integrates via MCP (Model Context Protocol) server to inject capabilities into Claude Code, Cursor, Copilot, and other agents. Security is enforced through static analysis in CI/CD, content hashing, path isolation, and Snyk scanning before publication—addressing the 13% vulnerability rate found in open marketplaces. Built as a monorepo (Nx, TypeScript 100%) with semantic versioning and immutable lockfiles.

About ai-agent-skills

wednesday-solutions/ai-agent-skills

Pre-configured agent skills for Vibe Coded projects. These skills provide AI coding assistants (Claude Code, Cursor, etc.) with specific guidelines for code quality and design standards.

Integrates deeply with Claude Code, Cursor, and Gemini CLI as configuration files (`.cursorrules`, `CLAUDE.md`, etc.) rather than standalone scripts, enabling AI assistants to enforce Wednesday Solutions' conventions without API calls. Core architecture uses SQLite graph databases (`graph.db`) built from AST parsing and git history to power deterministic codebase queries—enabling features like risk scoring, drift detection, and architecture-boundary validation without hallucination-prone LLM reads of raw source files. Covers the full lifecycle from greenfield planning (parallel Architect/PM/Security personas) through brownfield intelligence (structural queries, test generation, deployment checklists) and git discipline (conventional commits, PR automation, semantic fix queues).

Scores updated daily from GitHub, PyPI, and npm data. How scores work