write-specs
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a structured workflow for requirement elicitation with an explicit user approval stage, preventing unauthorized or automated implementation changes.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading repository code and documentation to gather facts for the SPEC creation. This presents a potential surface for instructions embedded in the codebase to influence the agent's output.\n
- Ingestion points: Local repository files, documentation, and code facts (SKILL.md, 'Fact Gathering').\n
- Boundary markers: The skill uses a structured 'Design Tree' and a fixed SPEC template which provides implicit structure, though no explicit 'ignore instructions' delimiters are applied to file content.\n
- Capability inventory: File system write access to the
.specs/directory and command execution via thecreate-issuesskill for GitHub integration.\n - Sanitization: There is no explicit sanitization or filtering logic described for the data read from the repository.
Audit Metadata