#4242nd of 54 in Agent and MCP starters
opengap
Git-repo file standard and CLI for defining portable AI agents
- Stars
- 3k
- License
- MIT
- Last commit
- Jul 2026
- Last release
- Jul 2026
- Language
- TypeScript
2,668 on 21 Apr 20262,986 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenGAP defines an agent as a git repo: agent.yaml and SOUL.md are required, with optional RULES.md, DUTIES.md, skills/, tools/, workflows/, memory/ and sub-agents. The opengap CLI (npm, Node >= 18) validates the definition against a spec and exports it to targets such as Claude Code, OpenAI Agents SDK, CrewAI, Gemini CLI, Cursor and plain system prompts. It also covers financial compliance checks (FINRA, SR 11-7) and segregation-of-duties validation.
Who it is for: Teams versioning agent prompts and policies in git, exporting to many frameworks
Strengths
- Exports one agent definition to about 15 targets, including Claude Code, OpenAI, CrewAI, Cursor and Gemini
- Only two required files, so a minimal agent is agent.yaml plus SOUL.md
- validate --compliance checks segregation-of-duties conflicts before deployment
- Agent history, diffs, branches and tags come from plain git
Weaknesses
- Spec is v0.1.0, so the format may still change
- Ports only identity layer; orchestration, tool execution and memory I/O stay in the framework
- Compliance features are aimed at US financial regulation, with little else covered
- Package was renamed twice (gitagent, gapman, opengap), which complicates searching docs and installs
What it needs
- TypeScript, Claude, OpenAI, Gemini, GitHub Models
- Needs Node.js >= 18, git
Quick answers
Is opengap open source?
Yes. Its code is on GitHub under the MIT license.
Also in Agent and MCP starters
See all 54| Rank | Project | Score |
|---|---|---|
| 1 | deepseek-harnessPlugin-based agent harness from DeepSeek with a local Web UI | 98 out of 100 |
| 2 | composioTypeScript and Python SDKs that give agents authenticated tools across 1000+ apps | 92 out of 100 |
| 3 | conductorDurable workflow engine that orchestrates microservices and LLM agents from JSON | 91 out of 100 |
| #4 | mcp-useTypeScript framework for building MCP servers with React Views and an Inspector | 87 out of 100 |
| #5 | a2uiProtocol and renderers for agents to stream native UI instead of text | 86 out of 100 |