come-again
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text via the
/come-againcommand, which represents an ingestion point for untrusted instructions. - Ingestion points: The text provided after the
/come-againcommand is used directly as input for the restatement task. - Boundary markers: There are no delimiters or specific instructions provided to isolate user input from the agent's restatement logic.
- Capability inventory: The skill contains strong negative constraints, explicitly prohibiting planning, file system modifications, worker execution, or merging.
- Sanitization: No sanitization, escaping, or validation of the user-provided text is described.
Audit Metadata