improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes codebase files that could contain malicious instructions.
- Ingestion points: Codebase navigation and exploration via the Agent tool (SKILL.md).
- Boundary markers: No delimiters or isolation instructions are provided when reading codebase content.
- Capability inventory: The skill can create GitHub issues using 'gh' and spawn sub-agents (SKILL.md).
- Sanitization: No validation or sanitization of codebase data is performed.
- [COMMAND_EXECUTION]: The skill executes shell commands to interact with GitHub for issue creation.
- Evidence: Step 7 instructions use 'gh issue create' (SKILL.md).
Audit Metadata