answer-buzzkill
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of pasted messages from third parties (colleagues).
- Ingestion points: The skill takes the colleague's message as input to find the core question and context.
- Boundary markers: The instructions do not specify specific delimiters for the untrusted input beyond "the input is their pasted message."
- Capability inventory: The skill utilizes read-only access to source code, git history, documentation (e.g., CONCEPTS.md, GLOSSARY.md), and issue trackers. Importantly, it contains explicit instructions forbidding any writing operations (edits, commits, ticket comments) or external messaging beyond the chat interface.
- Sanitization: No explicit sanitization or escaping of the interpolated message is described in the instructions.
- [SAFE]: The skill includes a reference to a YouTube video. This is a well-known service and the link is presented for informational context without any automated execution or download patterns.
Audit Metadata