What the reviewer found
A personal AGENTS.md plus docs for a CLI account-vault tool (caam) that backs up, activates and rotates local auth files for several AI CLIs (Claude, Codex, Gemini, Grok) into a local vault; it reads those config paths as its stated purpose, not to exfiltrate. The rm -rf mention is a defensive 'never run without written permission' rule and the HTML comment is a tool marker, both false; the unusual instruction to spoof a fake user-agent on outgoing curl requests is odd but not harmful to the installer.
credential-access— reads credentialsdual-use— a security tool that can be misused
What was read
The file as it ships in Dicklesworthstone/coding_agent_account_manager:
AGENTS.md
What the static scan said
The scan flagged 4things. The reviewer kept 1 and dismissed 3 as false.
P2Hidden instructions — false positiveAS1Reads agent configuration directories — realRMRecursive force delete — false positiveNETMakes network calls — false positive
How this review was made
Sonnet 5 read the files above on 7 September 2026 and answered three questions: is it dangerous to whoever installs it, is each scanner finding real, and what should the installer know. The verdict is bound to the file's hash; when the file changes, it is scanned afresh and reviewed again. A script that changes while the definition does not is not re-reviewed — that is a known gap. How the scan and the review work.