to-spec
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from both conversation history and the codebase to generate technical specifications, creating a surface for indirect prompt injection where malicious content in those sources could influence agent behavior.
- Ingestion points: The instructions explicitly direct the agent to ingest the "conversation" and the "codebase" (SKILL.md) as primary sources for synthesis.
- Boundary markers: The skill lacks explicit delimiters or instructions for the agent to ignore or sanitize embedded commands within the ingested codebase or conversation data.
- Capability inventory: The agent is authorized to read the codebase, "probe" production environments to measure live data/state, dispatch sub-agents, and publish results to an external "tracker."
- Sanitization: No sanitization, validation, or escaping mechanisms are defined for data ingested from external sources before it is used to drive the agent's "probing" actions or its final output to the tracker.
Audit Metadata