modular-decomposition
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, untrusted codebases as its primary function, creating an attack surface where malicious instructions embedded in the analyzed code could influence agent behavior.
- Ingestion points: The agent is instructed to scan directory structures and read source files from the user's repository across the entire five-pattern analysis pipeline (e.g., in
pattern-01-identify-and-size.md). - Boundary markers: The instructions lack specific guidance for the agent to use delimiters or 'ignore embedded instructions' prompts when reading repository content.
- Capability inventory: The skill methodology includes scanning and inventorying components but does not define its own network or file-write operations; it relies on the agent's native tools.
- Sanitization: No sanitization or filtering logic is provided to handle potentially malicious content within the analyzed files.
Audit Metadata