spec-writer
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected in the skill instructions or provided samples.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided file paths and text lists (SKILL.md, Input section). While it lacks explicit boundary markers or instructions to ignore embedded prompts in the ingested data, the risk is minimal as the output is restricted to generating architectural specifications in a markdown format saved to the user's Documents folder.
- Ingestion points: User-provided text and file paths processed in the Input section of SKILL.md.
- Boundary markers: Absent; the instructions do not explicitly warn the model to ignore instructions found within the material lists.
- Capability inventory: Uses Read, Write, and Bash tools as defined in the SKILL.md frontmatter.
- Sanitization: Absent; the model relies on its internal logic to parse and classify materials into standard CSI divisions.
- [COMMAND_EXECUTION]: The skill configuration allows access to the Bash tool; however, the instructions focus entirely on logical mapping and text generation for specifications. There are no instructions to execute arbitrary or dangerous shell commands.
Audit Metadata