security-audit-context-building
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Bash utilities including
find,cat, andgrepto perform local tech stack discovery and locate files potentially containing sensitive information (e.g.,.env,.yaml,.json). These commands are restricted to the local filesystem and serve the stated purpose of security auditing. - [PROMPT_INJECTION]: The skill processes untrusted content from the local codebase (e.g., package manifest files and git logs), creating a surface for indirect prompt injection.
- Ingestion points: Reads content from
package.json,.envfilenames, and repository history viagit log. - Boundary markers: No explicit delimiters are used to separate file content from agent instructions.
- Capability inventory: Uses
Bashtools (find, cat, grep, git) for context building. - Sanitization: No validation or sanitization of file content is performed prior to processing.
Audit Metadata