What it does
The publisher describes each round as starting with context, checking each path, and ending with an audit. No further details are given in the available information.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 14 Nov 2026 · its page on the source
spec-harness
Rounds of agent work under a brief: context to start, a check on each path, an audit at the end.
v0.26.0As shown on registry.modelcontextprotocol.io when we last read it