cria-techspec
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from PRD files (
tasks/prd-[feature-slug]/prd.md) and the local codebase to generate documentation. This creates an attack surface where malicious instructions embedded in these files could attempt to influence the agent's behavior during the research phase or when writing to the filesystem. - Ingestion points: PRD files (Step 2) and project source code (Step 3).
- Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore' instructions when reading untrusted content.
- Capability inventory: Web search (Step 4), interactive user questioning (Step 5), and file writing to the local directory (Step 8).
- Sanitization: Absent; the skill does not perform validation or sanitization on the ingested file content before it is used in the generation process.
Audit Metadata