wait-what
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and use information from external files (
CONTEXT.mdandCONTEXT-MAP.md) to guide its responses. - Ingestion points: The skill reads
CONTEXT.mdandCONTEXT-MAP.md(referenced inSKILL.md). - Boundary markers: None identified; the agent is instructed to use the 'ubiquitous language' from these files without explicit delimitation or warnings to ignore instructions embedded within them.
- Capability inventory: No dangerous capabilities such as network operations, file writes, or command execution were found in the skill files.
- Sanitization: There is no evidence of sanitization or filtering for the content ingested from the context files.
Audit Metadata