create-architectural-decision-record
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts untrusted user input to generate documentation files, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: Context, Decision, Alternatives, and Stakeholders inputs defined in SKILL.md.
- Boundary markers: The skill does not implement specific boundary markers or "ignore" instructions to prevent user-supplied content from potentially overriding the template structure.
- Capability inventory: The skill is capable of writing files to the /docs/adr/ directory as specified in the SKILL.md requirements.
- Sanitization: There is no evidence of sanitization or filtering of the user-provided input before it is written to the resulting markdown file.
Audit Metadata