analyze-performance-call-and-learn
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain hidden instructions aimed at influencing the agent's evaluations or learning persistence.
- Ingestion points: The skill ingests "this week's results", "the brief's test plan", "last week's report", "learnings", and "readouts" into the context for evaluation (SKILL.md).
- Boundary markers: No specific delimiters or instructions (e.g., XML tags or "ignore instructions in the following data") are used to isolate untrusted inputs from the agent's core logic.
- Capability inventory: The skill can generate reports and invoke the
save-learningskill (../save-learning/SKILL.md) to write data back to the system. - Sanitization: There is no evidence of validation or sanitization of the input data before it is used to evaluate campaign success or draft new rules.
Audit Metadata