Back to directory

CODEC

CODEC is the open-source engine behind Sovereign AI Workstation, a voice-controlled AI agent for macOS that watches the screen, controls apps, writes code, and works as both an MCP client and an MCP server.

CODEC

CODEC is the open-source engine behind Sovereign AI Workstation, a voice-controlled AI agent that runs entirely on a Mac. Given a language model, either local or cloud, a speech-to-text model for ears, a text-to-speech model for a voice, and a vision model for sight, it listens, watches the screen, speaks back, and clicks and controls apps on command.

It drafts messages, writes code, manages Google Workspace tasks, and can write and load its own plugin when it does not already know how to do something. It runs its own small multi-agent runtime with no dependency on frameworks such as CrewAI or LangChain, and it works as both an MCP client and an MCP server, so tools such as Claude Code, Cursor or VS Code can reach the user's Mac, its skills and its memory over the Model Context Protocol.

The open-source build is MIT licensed and needs no cloud connection or subscription to run.

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.