{"owner":"langchain-ai","repo":"open_deep_research","hasSkills":true,"hasMcp":false,"mcpConfig":null,"found":["CLAUDE.md"],"skills":{"CLAUDE.md":"# Open Deep Research Repository Overview\n\n## Project Description\nOpen Deep Research is a configurable, fully open-source deep research agent that works across multiple model providers, search tools, and MCP (Model Context Protocol) servers. It enables automated research with parallel processing and comprehensive report generation.\n\n## Repository Structure\n\n### Root Directory\n- `README.md` - Comprehensive project documentation with quickstart guide\n- `pyproject.toml` - Python project configuration and dependencies\n- `langgraph.json` - LangGraph configuration defining the main graph entry point\n- `uv.lock` - UV package manager lock file\n- `LICENSE` - MIT license\n- `.env.example` - Environment variables template (not tracked)\n\n### Core Implementation (`src/open_deep_research/`)\n- `deep_researcher.py` - Main LangGraph implementation (entry point: `deep_researcher`)\n- `configuration.py` - Configuration management and settings\n- `state.py` - Graph state definitions and data structures  \n- `prompts.py` - System prompts and prompt templates\n- `utils.py` - Utility functions and helpers\n- `files/` - Research output and example files\n\n### Legacy Implementations (`src/legacy/`)\nContains two earlier research implementations:\n- `graph.py` - Plan-and-execute workflow with human-in-the-loop\n- `multi_agent.py` - Supervisor-researcher multi-agent architecture\n- `legacy.md` - Documentation for legacy implementations\n- `CLAUDE.md` - Legacy-specific Claude instructions\n- `tests/` - Legacy-specific tests\n\n### Security (`src/security/`)\n- `auth.py` - Authentication handler for LangGraph deployment\n\n### Testing (`tests/`)\n- `run_evaluate.py` - Main evaluation script configured to run on deep research bench\n- `evaluators.py` - Specialized evaluation functions  \n- `prompts.py` - Evaluation prompts and criteria\n- `pairwise_evaluation.py` - Comparative evaluation tools\n- `supervisor_parallel_evaluation.py` - Multi-threaded evaluation\n\n### Examples (`examples/`)\n- `arxiv.md` - ArXiv research example\n- `pubmed.md` - PubMed research example\n- `inference-market.md` - Inference market analysis examples\n\n## Key Technologies\n- **LangGraph** - Workflow orchestration and graph execution\n- **LangChain** - LLM integration and tool calling\n- **Multiple LLM Providers** - OpenAI, Anthropic, Google, Groq, DeepSeek support\n- **Search APIs** - Tavily, OpenAI/Anthropic native search, DuckDuckGo, Exa\n- **MCP Servers** - Model Context Protocol for extended capabilities\n\n## Development Commands\n- `uvx langgraph dev` - Start development server with LangGraph Studio\n- `python tests/run_evaluate.py` - Run comprehensive evaluations\n- `ruff check` - Code linting\n- `mypy` - Type checking\n\n## Configuration\nAll settings configurable via:\n- Environment variables (`.env` file)\n- Web UI in LangGraph Studio\n- Direct configuration modification\n\nKey settings include model selection, search API choice, concurrency limits, and MCP server configurations."},"files":{"CLAUDE.md":"# Open Deep Research Repository Overview\n\n## Project Description\nOpen Deep Research is a configurable, fully open-source deep research agent that works across multiple model providers, search tools, and MCP (Model Context Protocol) servers. It enables automated research with parallel processing and comprehensive report generation.\n\n## Repository Structure\n\n### Root Directory\n- `README.md` - Comprehensive project documentation with quickstart guide\n- `pyproject.toml` - Python project configuration and dependencies\n- `langgraph.json` - LangGraph configuration defining the main graph entry point\n- `uv.lock` - UV package manager lock file\n- `LICENSE` - MIT license\n- `.env.example` - Environment variables template (not tracked)\n\n### Core Implementation (`src/open_deep_research/`)\n- `deep_researcher.py` - Main LangGraph implementation (entry point: `deep_researcher`)\n- `configuration.py` - Configuration management and settings\n- `state.py` - Graph state definitions and data structures  \n- `prompts.py` - System prompts and prompt templates\n- `utils.py` - Utility functions and helpers\n- `files/` - Research output and example files\n\n### Legacy Implementations (`src/legacy/`)\nContains two earlier research implementations:\n- `graph.py` - Plan-and-execute workflow with human-in-the-loop\n- `multi_agent.py` - Supervisor-researcher multi-agent architecture\n- `legacy.md` - Documentation for legacy implementations\n- `CLAUDE.md` - Legacy-specific Claude instructions\n- `tests/` - Legacy-specific tests\n\n### Security (`src/security/`)\n- `auth.py` - Authentication handler for LangGraph deployment\n\n### Testing (`tests/`)\n- `run_evaluate.py` - Main evaluation script configured to run on deep research bench\n- `evaluators.py` - Specialized evaluation functions  \n- `prompts.py` - Evaluation prompts and criteria\n- `pairwise_evaluation.py` - Comparative evaluation tools\n- `supervisor_parallel_evaluation.py` - Multi-threaded evaluation\n\n### Examples (`examples/`)\n- `arxiv.md` - ArXiv research example\n- `pubmed.md` - PubMed research example\n- `inference-market.md` - Inference market analysis examples\n\n## Key Technologies\n- **LangGraph** - Workflow orchestration and graph execution\n- **LangChain** - LLM integration and tool calling\n- **Multiple LLM Providers** - OpenAI, Anthropic, Google, Groq, DeepSeek support\n- **Search APIs** - Tavily, OpenAI/Anthropic native search, DuckDuckGo, Exa\n- **MCP Servers** - Model Context Protocol for extended capabilities\n\n## Development Commands\n- `uvx langgraph dev` - Start development server with LangGraph Studio\n- `python tests/run_evaluate.py` - Run comprehensive evaluations\n- `ruff check` - Code linting\n- `mypy` - Type checking\n\n## Configuration\nAll settings configurable via:\n- Environment variables (`.env` file)\n- Web UI in LangGraph Studio\n- Direct configuration modification\n\nKey settings include model selection, search API choice, concurrency limits, and MCP server configurations."},"items":[{"name":"CLAUDE.md","path":"CLAUDE.md","title":"CLAUDE.md","content":"# Open Deep Research Repository Overview\n\n## Project Description\nOpen Deep Research is a configurable, fully open-source deep research agent that works across multiple model providers, search tools, and MCP (Model Context Protocol) servers. It enables automated research with parallel processing and comprehensive report generation.\n\n## Repository Structure\n\n### Root Directory\n- `README.md` - Comprehensive project documentation with quickstart guide\n- `pyproject.toml` - Python project configuration and dependencies\n- `langgraph.json` - LangGraph configuration defining the main graph entry point\n- `uv.lock` - UV package manager lock file\n- `LICENSE` - MIT license\n- `.env.example` - Environment variables template (not tracked)\n\n### Core Implementation (`src/open_deep_research/`)\n- `deep_researcher.py` - Main LangGraph implementation (entry point: `deep_researcher`)\n- `configuration.py` - Configuration management and settings\n- `state.py` - Graph state definitions and data structures  \n- `prompts.py` - System prompts and prompt templates\n- `utils.py` - Utility functions and helpers\n- `files/` - Research output and example files\n\n### Legacy Implementations (`src/legacy/`)\nContains two earlier research implementations:\n- `graph.py` - Plan-and-execute workflow with human-in-the-loop\n- `multi_agent.py` - Supervisor-researcher multi-agent architecture\n- `legacy.md` - Documentation for legacy implementations\n- `CLAUDE.md` - Legacy-specific Claude instructions\n- `tests/` - Legacy-specific tests\n\n### Security (`src/security/`)\n- `auth.py` - Authentication handler for LangGraph deployment\n\n### Testing (`tests/`)\n- `run_evaluate.py` - Main evaluation script configured to run on deep research bench\n- `evaluators.py` - Specialized evaluation functions  \n- `prompts.py` - Evaluation prompts and criteria\n- `pairwise_evaluation.py` - Comparative evaluation tools\n- `supervisor_parallel_evaluation.py` - Multi-threaded evaluation\n\n### Examples (`examples/`)\n- `arxiv.md` - ArXiv research example\n- `pubmed.md` - PubMed research example\n- `inference-market.md` - Inference market analysis examples\n\n## Key Technologies\n- **LangGraph** - Workflow orchestration and graph execution\n- **LangChain** - LLM integration and tool calling\n- **Multiple LLM Providers** - OpenAI, Anthropic, Google, Groq, DeepSeek support\n- **Search APIs** - Tavily, OpenAI/Anthropic native search, DuckDuckGo, Exa\n- **MCP Servers** - Model Context Protocol for extended capabilities\n\n## Development Commands\n- `uvx langgraph dev` - Start development server with LangGraph Studio\n- `python tests/run_evaluate.py` - Run comprehensive evaluations\n- `ruff check` - Code linting\n- `mypy` - Type checking\n\n## Configuration\nAll settings configurable via:\n- Environment variables (`.env` file)\n- Web UI in LangGraph Studio\n- Direct configuration modification\n\nKey settings include model selection, search API choice, concurrency limits, and MCP server configurations.","category":"root","tokens":736}]}