code-to-prd
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external codebases to generate documentation, which creates a potential surface for indirect prompt injection if the analyzed files contain hidden instructions.
- Ingestion points:
scripts/codebase_analyzer.pyreads files recursively from a user-provided project path. - Boundary markers: The skill instructions do not specify the use of strict delimiters or specific "ignore instructions" warnings for the raw code data processed by the AI agent.
- Capability inventory: The skill utilizes
scripts/codebase_analyzer.pyfor file/directory reading andscripts/prd_scaffolder.pyfor local filesystem writes to generate the documentation skeleton. - Sanitization: The analyzer scripts use regular expression filtering to extract specific patterns like routes and API calls, but the AI is ultimately expected to interpret the broader business logic within the files.
- [COMMAND_EXECUTION]: The skill's workflow relies on the execution of two included Python scripts (
scripts/codebase_analyzer.pyandscripts/prd_scaffolder.py). These scripts are designed to run locally and use only Python standard libraries, which is consistent with the skill's stated purpose.
Audit Metadata