prd-to-engineering-spec
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external Product Requirements Documents (PRDs), which presents an attack surface where untrusted input could attempt to influence the agent's behavior during the specification process.\n
- Ingestion points: The skill processes PRD content during Phase 0 (Validation) and Phase 1 (Decomposition) as described in
SKILL.md.\n - Boundary markers: The provided validation scripts do not define specific delimiters or boundary markers for PRD data.\n
- Capability inventory: The skill utilizes shell scripts (
scripts/generate_spec_skeleton.sh,scripts/validate_spec.sh) for file system operations, including directory creation and content scanning.\n - Sanitization: Input sanitization is not explicitly performed on the PRD text, although the
validate_spec.shscript scans for placeholders and security patterns.\n- [COMMAND_EXECUTION]: The skill includes Bash scripts intended for project scaffolding and linting. These scripts perform standard file system operations such as creating directories and searching for text patterns using grep. (Evidence:scripts/generate_spec_skeleton.sh,scripts/validate_spec.sh).
Audit Metadata