code-security-audit
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external codebases, creating a surface for indirect prompt injection if the source code contains malicious instructions.
- Ingestion points: The agent reads files from a user-specified codebase path (SKILL.md, references/audit_process.md).
- Boundary markers: The skill does not provide instructions to wrap file content in delimiters or warn the agent to ignore instructions embedded in the code.
- Capability inventory: The agent uses file search and read capabilities to perform the audit.
- Sanitization: No explicit sanitization or filtering of the ingested code content is described.
- Remediation: Implement explicit boundary markers (e.g., XML tags or clear separators) when providing file content to the LLM and include instructions to disregard any natural language commands found within the source code.
Audit Metadata