ClawAudit verdict
devlog
The skill seems to be designed for generating narrative blog posts from AI coding session transcripts, which is a legitimate and safe task.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (4)
Accesses sensitive environment variables
examples/builders-log.md · code · process.env.JWT_SECRET
Instructs covert action — may act without user awareness
examples/technical.md · prose · downgraded · silently
Popular HTTP library — network access
references/blog-writing-guide.md · code · Got
Python os.environ.get — reads environment variable
references/platforms/opencode/list-sessions.sh · prose · downgraded · os.environ.get(
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.