wait-what
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to read and incorporate content from
GLOSSARY.mdandGLOSSARY-MAP.md. - Ingestion points: Local file system access for
GLOSSARY.mdandGLOSSARY-MAP.mdwithin the skill repository. - Boundary markers: The prompt does not provide delimiters to separate the ingested glossary data from the instructions given to the agent.
- Capability inventory: The skill has no executable capabilities, tool access, network permissions, or file-writing logic. It is defined as a non-invocable prompt template.
- Sanitization: No validation or sanitization is applied to the content of the glossary files before processing.
Audit Metadata