security-app-websocket
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external project files, WebSocket frame schemas, and message types, which creates an ingestion point for untrusted data into the agent's context.
- Ingestion points: Reads project codebases, endpoint inventories, and potentially message logs during security reviews (Step 1, Step 10).
- Boundary markers: No explicit delimiters or boundary markers (like XML tags or specific 'ignore' blocks) are specified for the interpolation of the analyzed code into the agent's prompt.
- Capability inventory: The skill utilizes
Bash,Write, andEdittools to perform its tasks and report findings. - Sanitization: The skill includes robust sanitization rules via
sensitive_data_rules(forbidding the reading of.env,secrets/, and.pemfiles) andredact_on_output(enforcing redaction of session tokens and PII). - [SAFE]: The skill follows established security standards (OWASP WebSocket Security Cheat Sheet, ASVS 5.0) and uses a specialized SDK (
appsec-sdk) for reporting findings rather than direct file manipulation in sensitive paths, which is a defensive design pattern.
Audit Metadata