code-naming-auditor
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by reading code files from a user-defined directory to perform naming audits. There is a theoretical risk that malicious instructions embedded in the analyzed code (e.g., within comments or documentation strings) could influence the agent's behavior during the analysis or refactoring phase.
- Ingestion points: The skill reads files from directories specified by the user in Step 2 (Define the Audit Scope).
- Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers when reading file content.
- Capability inventory: The skill has the capability to read file content and use a
replacetool to modify files. - Sanitization: No explicit sanitization or filtering of the read content is mentioned in the workflow.
Audit Metadata