Back to directory

SourceWeft

A self-hostable, open-source AI workstation that runs multiple agents and sub-agents against your own models, knowledge and tools, with a skill library, MCP integrations and sandboxed execution.

SourceWeft

SourceWeft is an open-source workstation that brings conversations, knowledge, working files and task execution together in one place. Users describe a task, agents break it down and use the right tools, and the results feed into further tasks.

It supports multi-agent collaboration with delegated sub-agents, a skill library covering presentations, HTML pages, images, meeting summaries and learning material, MCP integrations plus web search and extraction, and sandboxed execution of approved commands over files. Users bring their own model provider API keys, and it can index PDFs, web pages, notes, YouTube, Notion, Google Drive, Gmail and Slack as sources with citations.

It runs through a web app or desktop clients for macOS, Windows and Linux, can be self-hosted for a team, and supports resuming saved conversations across devices.

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.