ClawAudit verdict
testtau
The skill provides disposable email inboxes and webhook capture for testing.
โ Flagged for review โ coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis โ not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Permission integrity
network_out
Findings (1)
Instruction-prose smuggling shape detected: collects a sensitive target ("API key") and emits it outward ("send"). Phrased as prose with no trigger tokens โ a semantic prompt-injection / data-exfil pattern the syntactic scanners can't see. Final tier capped at Caution; review the instructions before installing.
SKILL.md ยท - Public mail: send to `<inbox>@mail.testtau.com`, inspect at `https://mail.testtau.com/i/<inbox>`. - Public hooks: send to `https://hook.testtau.com/<hookId>`,
Permissions & capabilities
No declared permissions โ minimal attack surface.
network_outcredential_accessnetwork_in Is this flag fair?
Thanks โ recorded.