Back to directory

DocsGPT

A private AI platform for building agents, assistants and enterprise search, with an agent builder, document analysis across many file formats, deep research tools and API connectivity to external tools.

DocsGPT

DocsGPT is an open-source AI platform for building agents and assistants for enterprise search, with an Agent Builder, deep research tools, and document analysis covering PDF, Office, web content and audio files.

It reads a wide range of formats, including PDF, DOCX, CSV, XLSX, EPUB, HTML, JSON, PPTX, images and audio files such as MP3 and WAV, and can transcribe voice input or meeting recordings into searchable knowledge. It ingests content from URLs, sitemaps, Reddit, GitHub and web crawlers, and answers with source citations. Agents can call external APIs and tools, and pre-built integrations include chat widgets and Discord and Telegram bots.

It works with major hosted LLMs such as OpenAI, Google and Anthropic models, as well as local models through Ollama and llama.cpp, and can be deployed privately with Kubernetes support for enterprise use. It is written in Python.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.