A platform-agnostic multi-agent AI framework built as experiment and based on LangGraph and LiteLLM. Orchid lets you define AI agents via YAML configuration, orchestrate them with a supervisor, connect external tools via MCP servers, and augment responses with hierarchical RAG, all without writing agent code.
npx skills add gadz82/orchidA platform-agnostic multi-agent AI framework built as experiment and based on LangGraph and LiteLLM. Orchid lets you define AI agents via YAML configuration, orchestrate them with a supervisor, connect external tools via MCP servers, and augment responses with hierarchical RAG, all without writing agent code.
Source: https://github.com/gadz82/orchid
Discovered during the daily awesomeskills.dev agent-skill hunt.
AI skill documentation repo with reusable documentation assets. Documented structure of agent skills: SKILL.md, bundled references, scripts, examples, and boundaries.
21 custom Claude Code skills for content creation, PM workflows, and AI automation
A demo site to test out various Claude skills.
Repo-native memory layer for Claude Code, Cursor, and Codex. Local-first markdown ontology + MCP tools so AI coding agents keep codebase context.
Agent skills for DefProd — user story alignment, codebase analysis, test generation, drift detection
Codex skill and Python CLI for generating Chinese graduation thesis drafts