wait-what
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to ingest and use language from local repository files (
CONTEXT.mdandCONTEXT-MAP.md). - Ingestion points: Content is read from
CONTEXT.mdandCONTEXT-MAP.mdlocated in the repository. - Boundary markers: The skill lacks explicit delimiters or instructions for the agent to ignore potentially malicious embedded commands within these external context files.
- Capability inventory: The skill has
disable-model-invocation: trueset in its frontmatter and does not include any scripts with subprocess, network, or file-write capabilities. - Sanitization: There is no evidence of content validation or sanitization for the data retrieved from the context files.
Audit Metadata