$ show sickn33/agentic-awesome-skills
Agentic Awesome Skills is a curated marketplace infrastructure for AI coding agent skills—instructional playbooks (SKILL.md files) with accompanying scripts and assets. The repository provides a multi-platform distribution system supporting Claude Code, Cursor, Gemini CLI, Codex CLI, Kiro, and others. It includes a React+Vite web application for browsing, searching, and installing skills; a CLI and MCP server for local agent integration; and a comprehensive validation, security scanning, and release pipeline. The codebase implements a transaction engine with deterministic content addressing (canonical JSON + SHA-256), durable state management, and schema-validated skill installation, forming a robust foundation for agent-owned tool selection and workflow planning. While the core infrastructure (AAS v1 library, MCP server, catalog search) demonstrates high craftsmanship and reusability, the repository is hampered by massive file duplication: identical skill scripts and governance modules appear under `skills/`, `plugins/agentic-awesome-skills/`, and `plugins/agentic-awesome-skills-claude/`, creating synchronization risk. Some skills contain hardcoded paths or incomplete boilerplate. The repository is actively maintained with CI/CD, automated audits, and security scrutiny, though the sheer volume of 2,474+ skills imposes ongoing maintenance burden. Overall, it is a well-engineered, production-quality skill distribution platform with notable reusable design patterns.
The LLM identified 4 files worth highlighting for reuse:
Implements a durable managed state system with integrity checks: atomic writes via tempfile+rename, SHA-256 digest verification, schema validation on load, and corruption detection. Decouples state persistence from business logic. Reusable for any agent or CLI tool that needs crash-safe, tamper-evident configuration or progress state.
Produces deterministic JSON serialization with sorted keys, no whitespace, and consistent number formatting. Combined with SHA-256 hashing, it enables content-addressable storage and idempotent caching. Reusable in any system requiring reproducible JSON digests, e.g., for build artifact verification or ledger entries.
Full MCP (Model Context Protocol) server implementation over stdio transport, with tool registration, strict JSON validation, error codes, and prompt generation. Isolated from the rest of the AAS library via dependency injection. Reusable as a template for building MCP servers for agent tool execution, requiring minimal modification.
Provides a stateless catalog search engine supporting three modes (any, all, fuzzy) with token normalization, shared alias expansion (e.g., 'a11y' ↔ 'accessibility'), and literal word matching. Returns deterministic, scored results. Reusable in any front-end or back-end search scenario where curated, alias-aware text matching is needed.