ClawAudit verdict
turnstile-spin
Cloudflare Turnstile integration wizard using the CF API; all API calls target the user own Cloudflare account with confirmation before irreversible steps.
⚠ 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.
Permission integrity
network_out
Findings (8)
<script> tag in markdown — potential code injection
references/astro.md · code · <script
Possible hardcoded credential
scripts/auth-probe.sh · prose · downgraded · token="${CLOUDFLARE_API_TOKEN:-}
Pipe to python — executes piped content as Python code
scripts/auth-probe.sh · prose · downgraded · | python3
Instructs covert action — may act without user awareness
SKILL.md · prose · downgraded · silently
References child_process — can spawn system processes
templates/worker/test/deploy.test.ts · prose · downgraded · child_process
Makes HTTP request to external URL
references/astro.md · code · fetch("https://
Popular HTTP library — network access
scripts/validate.sh · prose · downgraded · got
Accesses sensitive environment variables
templates/worker/test/deploy.test.ts · prose · downgraded · process.env.CLOUDFLARE_API_TOKEN
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B/D/F). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_outnetwork_in Is this flag fair?
Thanks — recorded.