Skill Seekers is a Python 3.10+ command-line tool and MCP server that turns 18 source types — documentation sites, GitHub repositories, PDFs, Word and EPUB files, notebooks, OpenAPI specs, man pages, and video — into structured knowledge assets packaged for Claude AI skills, RAG pipelines such as LangChain, LlamaIndex, and Pinecone, and AI coding assistants including Cursor, Windsurf, and Cline.

What it is

Skill Seekers lives in the Claude AI and Model Context Protocol (MCP) ecosystem. It is distributed on PyPI as the skill-seekers package under the MIT licence, currently at version 3.9.0, with a hosted presence at skillseekersweb.com. The project describes itself as the data layer for AI systems: a preparation stage that reads raw sources once, normalises them into a knowledge asset, and exports that asset to 22 different targets. Topics on the repository — ast-parser, conflict-detection, code-analysis, documentation-generator, github-scraper, mcp, claude-skills — describe the mechanics behind that pipeline.

The concrete problem it solves is per-source integration work. Instead of writing a separate scraper or converter for every documentation site, repository, PDF, or slide deck a team needs, a single command covers GitHub repositories, local codebases, PDF, DOCX, EPUB, Jupyter notebooks, OpenAPI/Swagger, PowerPoint, AsciiDoc, local HTML, RSS/Atom, man pages, and video. What it replaces is the hand-built pipeline between a source of truth and the AI system that consumes it: skill-seekers create ingests the source, and skill-seekers package writes the finished artefact for a chosen target.

Key capabilities

• skill-seekers create ingests all 18 source types through one command, from facebook/react and local codebases to manual.pdf, report.docx, book.epub, notebook.ipynb, openapi.yaml, presentation.pptx, guide.adoc, page.html, feed.rss, and curl.1 man pages.
• skill-seekers package output/django --target claude produces output/django-claude.zip, one of 22 export targets spanning Claude, Gemini, and OpenAI skills, RAG stacks, and coding assistants.
• skill-seekers scan ./my-react-app --out ./configs/scanned/ has an AI agent read manifests, README, Dockerfile/CI, and sampled source imports, then emits one config per detected framework — react.json, vite.json, tailwind.json, jest.json — plus a project codebase JSON file.
• MCP integration exposes 40 tools to MCP-capable clients.
• Automatic conflict detection runs across ingested sources, with AST parsing and code analysis supporting repository and codebase inputs.
• Enhancement runs through a pluggable agent: the default is claude, and --agent kimi or --agent-cmd "my-custom-agent run" selects another.
• Generated configs can be published to the community registry at yusufkaraaslan/skill-seekers-configs, and Atlas Cloud is supported as a packaging target via --target atlas with ATLAS_API_KEY.

Who uses it and how

• Documentation and knowledge-base teams converting vendor documentation sites, such as the Django docs, into Claude skills without maintaining their own scraper.
• RAG engineers preparing source material for LangChain, LlamaIndex, or Pinecone ingestion, using the tool as the data-preparation step ahead of retrieval.
• Developers working in Cursor, Windsurf, or Cline who want framework- and codebase-scoped context, using scan to detect React, Vite, Tailwind, and Jest configurations across a project.
• Teams handling mixed formats — PDFs, Word documents, EPUB, notebooks, man pages, and video — where video ingestion requires the skill-seekers[video] extra and a YouTube, Vimeo, or local file source.
• Config authors who publish presets back to the community registry rather than keeping detection rules private.

Getting started

Install from PyPI with pip install skill-seekers, then run skill-seekers create followed by skill-seekers package to produce a target archive. A hosted option is available at skillseekersweb.com, and the README is translated into twelve additional languages.

How it compares

The similar tools named in the facts sit on either side of it rather than beside it: Claude, Gemini, and OpenAI skills are output formats, and LangChain, LlamaIndex, Pinecone, Cursor, Windsurf, and Cline are consumers of the packaged asset. Skill Seekers positions itself as the preparation layer that feeds those systems, not a replacement for any of them, and it is permissively licensed under MIT rather than sold as a hosted product.

When to use it — and when not

A self-hoster needs a Python 3.10+ environment, and video ingestion adds the optional skill-seekers[video] dependency; the scan path depends on an external AI agent or a custom command being available. Teams that want a fully managed service with no runtime to maintain, or that need conflict resolution rather than conflict detection, should verify fit first — the README describes detection but gives no detail on how disagreements are resolved, and the repository carries 59 open issues.

# 1. Install
pip install skill-seekers

# 2. Create a skill from any source
skill-seekers create https://docs.djangoproject.com/

# 3. Package it for your AI platform
skill-seekers package output/django --target claude

# Pick a different AI agent for enhancement (default: claude)
skill-seekers create https://docs.djangoproject.com/ --agent kimi
skill-seekers create https://docs.djangoproject.com/ --agent-cmd "my-custom-agent run"

skill-seekers scan ./my-react-app --out ./configs/scanned/
# → react.json, vite.json, tailwind.json, jest.json, my-react-app-codebase.json

skill-seekers create ./configs/scanned/react.json

skill-seekers create facebook/react            # GitHub repository
skill-seekers create ./my-project              # Local codebase
skill-seekers create manual.pdf                # PDF
skill-seekers create report.docx               # Word
skill-seekers create book.epub                 # EPUB
skill-seekers create notebook.ipynb            # Jupyter
skill-seekers create openapi.yaml              # OpenAPI/Swagger
skill-seekers create presentation.pptx         # PowerPoint
skill-seekers create guide.adoc                # AsciiDoc
skill-seekers create page.html                 # Local HTML (or a whole dir)
skill-seekers create feed.rss                  # RSS/Atom
skill-seekers create curl.1                    # Man page

# Video (YouTube, Vimeo, or local — needs skill-seekers[video])
skill-seekers create --video-url https://www.youtube.com/watch?v=... --name mytutorial
skill-seekers create --setup                   # auto-install GPU-aware visual deps

skill-seekers create --space-key TEAM --name wiki               # Confluence
skill-seekers create --database-id ... --name docs              # Notion
skill-seekers create --chat-export-path ./slack-export --name team-chat  # Slack/Discord

pip install skill-seekers              # Core: scraping, GitHub, PDF, packaging
pip install skill-seekers[all-llms]    # + every LLM platform
pip install skill-seekers[mcp]         # + MCP server
pip install skill-seekers[all]         # Everything