makes your agent (or you) prove a claim before anyone spends time on it. feasibility, optimality, brainstorming, agent-to-agent review. built from real debates.
Let your AI agent talk directly to someone else's. One pairing code, end-to-end encrypted, no public IP. MCP server for Claude Code, Codex, Cursor, Gemini CLI and grok.
Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya
Source
GitHub
Language
Python
Last activity
Oct 16, 2026
Stars763Measured · Oct 18, 2026
Momentum70.87 days · Measured · +17.2% relative change
Audit what your coding agents silently ignore: Claude Code's CLAUDE.md, skills, sub-agents, hooks, settings, permissions and MCP, plus Cursor rules, Copilot instructions and AGENTS.md. Python, no dependencies, no network.
AROP — the open, vendor-neutral Agent Runtime Operations Protocol: publish, register, authorize, dispatch, stream, trace, audit, and retire AI agents across any framework. Complements MCP & A2A with a production operations layer. JSON Schema / OpenAPI / AsyncAPI, Go·Python·TS SDKs.
Pass the prompt, not the payload. Hand the next AI agent a short, checksummed reference to a prompt; the receiver gets the exact text or an error. Vault backends: SQLite, S3, GitHub Gist. Optional HMAC signing, CLI, zero dependencies. Not encryption. By Data.PyCap.
Open-source packages for .agt agent names: @agtnames/resolver, @agtnames/mcp (MCP server, com.agtnames/agt), @agtnames/countersign, the agt-names skill and the manifest spec