v3.0.0 release langchain llamaindex cursor claude

Skill Seekers v3.0.0: The Universal Intelligence Platform

Transform docs, GitHub repos, PDFs, and codebases into structured knowledge for any AI system. 16 output formats. 1,852 tests. One tool for LangChain, LlamaIndex, Cursor, Claude, and more.

Skill Seekers Team •

Skill Seekers v3.0.0: The Universal Intelligence Platform

TL;DR

  • 🚀 16 output formats (was 4 in v2.x)
  • 🛠️ 26 MCP tools (was 9)
  • ✅ 1,852 tests passing (was 700+)
  • ☁️ Cloud storage support (S3, GCS, Azure)
  • 🔄 CI/CD ready (GitHub Action + Docker)
  • 🎮 Godot game engine support with signal flow analysis
  • 🌐 27+ programming languages (7 new)
  • 📚 4 input sources: Docs, GitHub repos, PDFs, Local codebases
pip install skill-seekers
skill-seekers scrape --config react.json

The Problem We’re Solving

Every AI project needs data preprocessing:

  • RAG pipelines: “Scrape these docs/repos/PDFs, chunk them, embed them…”
  • AI coding tools: “I wish Cursor knew this framework/API…”
  • Claude skills: “Convert this codebase into a skill”

70% of RAG development time is spent on data preprocessing. Everyone rebuilds the same infrastructure. Stop rebuilding. Start using.

The Solution: Universal Preprocessor

Skill Seekers v3.0.0 transforms docs, GitHub repos, PDFs, and local codebases into structured knowledge for any AI system:

For RAG Pipelines

# From documentation
skill-seekers scrape --format langchain --config react.json

# From GitHub repository
skill-seekers scrape --format langchain --github https://github.com/user/repo

# From PDF files
skill-seekers scrape --format langchain --pdf ./manual.pdf

# From local codebase
skill-seekers analyze --directory ./my-project --format langchain

For AI Coding Assistants

# Works with any source - docs, repos, or codebases
skill-seekers scrape --target claude --config react.json
cp output/react-claude/.cursorrules ./

# Windsurf, Cline, Continue.dev - same process

For Claude AI

skill-seekers install --config react.json
# Auto-fetches, scrapes, enhances, packages, uploads

What’s New in v3.0.0

4 Input Sources

SourceCommandUse Case
Documentationscrape --configFramework docs, APIs, guides
GitHub Reposscrape --githubOpen source libraries, tools
PDF Filesscrape --pdfManuals, research papers, books
Local Codebasesanalyze --directoryYour own projects, game engines

16 Platform Adaptors

CategoryPlatformsCommand
RAG/VectorsLangChain, LlamaIndex, Chroma, FAISS, Haystack, Qdrant, Weaviate--format <name>
AI PlatformsClaude, Gemini, OpenAI--target <name>
AI CodingCursor, Windsurf, Cline, Continue.dev--target claude
GenericMarkdown--target markdown

26 MCP Tools

Your AI agent can now prepare its own knowledge:

  • Config tools (3): generate_config, list_configs, validate_config
  • Scraping tools (8): estimate_pages, scrape_docs, scrape_github, scrape_pdf, scrape_codebase, detect_patterns, extract_test_examples, build_how_to_guides
  • Packaging tools (4): package_skill, upload_skill, enhance_skill, install_skill
  • Source tools (5): fetch_config, submit_config, add/remove_config_source, list_config_sources
  • Splitting tools (2): split_config, generate_router
  • Vector DB tools (4): export_to_weaviate, export_to_chroma, export_to_faiss, export_to_qdrant

Cloud Storage

Upload skills directly to cloud storage:

# AWS S3
skill-seekers cloud upload output/react/ --provider s3 --bucket my-bucket

# Google Cloud Storage
skill-seekers cloud upload output/react/ --provider gcs --bucket my-bucket

# Azure Blob Storage
skill-seekers cloud upload output/react/ --provider azure --container my-container

CI/CD Ready

GitHub Action:

- uses: skill-seekers/action@v1
  with:
    config: configs/react.json
    format: langchain

Docker:

docker run -v $(pwd):/data skill-seekers:latest scrape --config /data/config.json

Godot Game Engine Support

Full Godot 4.x analysis with signal flow detection:

skill-seekers analyze --directory ./my-godot-game --comprehensive

Detects:

  • Signal declarations and connections
  • Event patterns (EventBus, Observer, Event Chains)
  • GDScript test extraction (GUT, gdUnit4)

Extended Language Support

7 New Languages: Dart, Scala, SCSS/SASS, Elixir, Lua, Perl

Total: 27+ programming languages supported

Production Quality

  • ✅ 1,852 tests across 100 test files
  • ✅ 58,512 lines of Python code
  • ✅ 80+ documentation files
  • ✅ 12 example projects for every integration

Quick Start

# Install
pip install skill-seekers

# Create a config
skill-seekers config --wizard

# Or use a preset
skill-seekers scrape --config configs/react.json

# Package for your platform
skill-seekers package output/react/ --target langchain

Migration from v2.x

v3.0.0 is fully backward compatible. All v2.x configs and commands work unchanged. New features are additive.

Ready to transform your data into AI knowledge?

pip install skill-seekers

The universal preprocessor for AI systems.

← Back to Blog