Skill Seekers

Skill Seekers prepares documentation and code for AI assistants. It's for developers and teams who want reusable skills, coding context or retrieval-augmented generation (RAG) data without preparing the same sources separately for each tool.

• Combined sources: Build a knowledge asset from documentation sites, public or private GitHub repositories, local codebases, PDFs, notebooks and workplace content such as Confluence or Notion.
• Code and documentation checks: Extract usage examples from tests, identify design patterns and architecture, and flag disagreements between written guidance and code.
• Document and video extraction: Read JavaScript-rendered documentation and scanned PDFs, extract on-screen code from videos, and use local Whisper transcription when subtitles aren't available.
• Reusable exports: Produce skills for Claude, Gemini and OpenAI, context for Cursor and Windsurf, and data for LangChain, LlamaIndex, Chroma, FAISS and Qdrant without collecting the sources again.

The Python software runs locally and is open source under the MIT license. AI enhancement can use locally installed agents such as Claude Code, Codex or OpenCode, or APIs from Anthropic, Google and OpenAI. API enhancement sends content to the chosen provider.

An MCP server lets compatible assistants prepare knowledge directly. Optional search indexing helps agents find passages within generated references. Documentation change detection and incremental updates keep generated assets aligned with their inputs.