speckit-constitution
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: No malicious behavior detected. The skill is restricted to local file operations for documentation purposes.
- [NO_CODE]: The skill contains only natural language instructions and does not ship with any executable scripts or binary files.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data. Ingestion points: User arguments and project documentation (README.md, docs/quickstart.md). Boundary markers: Absent; no specific delimiters isolate external content from the skill's logic. Capability inventory: File reads and writes limited to the local project environment. Sanitization: The skill performs validation to ensure all template tokens are replaced before final output.
Audit Metadata