grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's instructions and configuration do not contain malicious patterns, such as data exfiltration, obfuscation, or unauthorized command execution. It uses local file paths consistent with the author's namespace for storing decision records.
- [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing external design plans and translating them into persistent documentation, which represents a potential attack surface.
- Ingestion points: User-supplied design plans and interview responses are processed to generate records.
- Boundary markers: The skill includes logic to "never turn guesses into accepted decisions" and to ask one decision at a time, which acts as a verification step, though it lacks explicit prompt delimiters for the ingested plan data.
- Capability inventory: The skill is limited to reading local files and writing markdown content to
CONTEXT.mdand the.jimmy/adr/directory. - Sanitization: The instructions do not specify explicit sanitization or escaping of the plan content before it is written to architectural records.
Audit Metadata