first-principles-coach
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection via external data ingestion.\n
- Ingestion points: Phase 1 of
SKILL.mdinstructs the agent to useWebFetchto read content from a user-provided URL to establish business context.\n - Boundary markers: The skill instructions do not specify the use of delimiters or clear separation between system instructions and untrusted external data.\n
- Capability inventory: The skill requests the ability to read external data. No high-risk capabilities such as arbitrary command execution, file writing, or network exfiltration were identified in the skill's own logic.\n
- Sanitization: No logic is provided to sanitize or validate the content fetched from external URLs before it is processed by the agent.
Audit Metadata