Back to directory

Pragma

A cross-harness platform for building portable teams of AI agents, called Missions, that combine models, tools, shared context and memory once and run unchanged in Codex, Claude Code, Pragma's own desktop app, or an embedding SDK.

Pragma

Pragma treats an agent team and its working method, defined as Experts, ExpertTeams, Flows, tools, Skills, shared context, memory, permissions and human checkpoints, as one portable unit described in a YAML DSL, rather than a single prompt tied to one chat product.

A team can be built and run through Pragma Desktop, run from a terminal through its CLI, or embedded into another application through its SDK, and the same team can coordinate different models and harnesses across the steps of one mission without losing decisions, artifacts or accumulated experience. Cross-harness memory accumulates across missions into stable knowledge and reusable skills.

The current release is a macOS desktop preview for Apple Silicon and Intel, distributed through GitHub Releases; the maintainers note it should be verified as coming from the official release before installing.

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.