ClawAudit verdict
clawSafe
clawsafe
Local security middleware that scans input for prompt injection, XSS, SQL injection, and other threats; operates entirely locally as a detection library with no network calls.
⚠ 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 (7)
Prompt injection — tries to override agent instructions
README.md · code · Ignore previous instructions
<script> tag in markdown — potential code injection
SKILL.md · prose · downgraded · <script>
Possible hardcoded credential
detector.js · prose · downgraded · apiKey = "sk-1234567890abcdefghij
Uses eval() — can execute arbitrary code
detector.js · prose · downgraded · eval(
HTTP request to bare IP address — common in malicious payloads
detector.js · prose · downgraded · http://127.0.0.1
Gopher protocol — SSRF attack vector
rules/web/ssrf.json · prose · downgraded · gopher://
Accesses sensitive environment variables
detector.js · prose · downgraded · process.env.SECRET_KEY
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.