security-audit-owasp-top-10
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill analyzes untrusted external codebases, creating a surface where malicious instructions embedded in comments or code could influence the agent's behavior.
- Ingestion points: Phase 1 and Phase 3 involve the use of
Glob,Grep, andReadtools to ingest arbitrary files from the user's project. - Boundary markers: The workflow does not specify the use of delimiters or 'ignore' instructions to distinguish code content from system instructions during semantic analysis.
- Capability inventory: The skill utilizes
Glob,Grep,Read, andWritecapabilities across all project files. - Sanitization: There is no evidence of sanitization or escaping of the code content before it is processed by the LLM for analysis.
- Data Exposure (SAFE): While the skill actively searches for sensitive files (e.g.,
.env, credentials, PII), this behavior is strictly aligned with the primary purpose of a security audit. No evidence of unauthorized network exfiltration or credential hardcoding was found.
Audit Metadata