ask-matt
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines workflows that ingest untrusted data from external sources, creating a potential attack surface.
- Ingestion points: Processes such as
/grill-with-docs,/triage, and/researchare designed to read from repository documentation (CONTEXT.md, ADRs), bug reports, and primary external sources. - Boundary markers: The routing instructions do not specify the use of explicit delimiters or instructions to ignore embedded commands within the ingested content.
- Capability inventory: The workflows described include capabilities for file modification, code implementation, and tool execution (e.g.,
gh-stack,git). - Sanitization: There is no mention of sanitization, validation, or filtering of the external data before it is processed by the agent.
Audit Metadata