Coding agents like Claude Code, Cursor and Codex now run commands, read files and call tools with our own permissions. After a long session, most of us couldn't say exactly what happened: the transcript is a conversation, not a record.
Curious how others handle this:
- Do you review what the agent ran, or just the final diff?
- Has an agent ever done something you only noticed later?
Agentmetry is a local-first endpoint sensor for AI coding agents. It records each tool call Claude Code, Cursor, Codex and others make, in a hash-chained trail on the laptop, and can forward it to the SIEM you already run.
The difference is the MCP listing. It hashes tools/list per tool and records when it moves while the config stays the same: the MCP rug pull (ATLAS AML.T0109). It stores the hash, not the text.
Apache-2.0. No install needed: uvx agentmetry demo. Sees only agents with hooks.
