handoff
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes conversation history to generate a document that will be used as input for a subsequent agent session. This creates a multi-step chain where content from the original conversation could influence the behavior of the next agent.
- Ingestion point: Conversation history processed in
SKILL.md. - Capability inventory: Writing files to the operating system's temporary directory.
- Boundary markers: No explicit boundary markers or delimiters for the ingested history are specified.
- Sanitization: The instructions explicitly require redacting sensitive information (API keys, passwords, PII) before saving.- [COMMAND_EXECUTION]: The skill directs the agent to write a file to the user's OS temporary directory, which is outside the project workspace. This is an intended functionality for creating cross-session handoff documents.- [NO_CODE]: The skill consists only of configuration files and instructions, with no executable scripts or external dependencies.
Audit Metadata