CHAP now speaks MCP
The boundary lines between human and digital intelligence collaboration are often murky.
And in regulated environments murky is never good enough.
Organisations orchestrating human-agent teams require an audit trail of approvals and overrides. This same evidence is also required to learn from the expert judgements your people make when correcting and overriding agents.
CHAP – our open-source Collaborative Human-Agent Protocol – gives you one standard way to capture it.
Every approval and override becomes a verifiable, replayable record. And because each override carries the reason behind it, that same trail becomes the data you use to make the agents better.
And CHAP now speaks MCP.
The reference coordinator ships as an MCP server – wire it into Claude Desktop, Claude Code, Cursor, or any MCP client and run it in plain language, with the audit trail building underneath.
Open standard, royalty-free.
Keep reading:
Including the Responsible Human
CHAP technical report (opens in a new tab)
