discover-event-surfaces
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill ingests a
change_briefYAML provided by the user and reads local source code files referenced in thesurfaces.componentsandfile_summary_mapfields. - Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" protocols when reading the content of the target source files.
- Capability inventory: The skill has the capability to read arbitrary local files based on path inputs and calls tools to fetch project context and event taxonomies from Amplitude.
- Sanitization: There is no explicit sanitization or validation logic described for the content read from the source code before it is processed by the LLM.
Audit Metadata