checkpoint
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (multi-step chain). It processes the current conversation history—which may contain untrusted data from external sources or previous inputs—to generate a summary that will guide the behavior of a future agent.
- Ingestion points: Processes the entirety of the current conversation history to create the handoff document.
- Boundary markers: No specific delimiters or safety instructions are provided to the agent to ignore potentially malicious instructions embedded in the conversation being summarized.
- Capability inventory: The skill utilizes shell commands (
mktemp), file reading, and file writing capabilities. - Sanitization: There is no evidence of sanitization or filtering of the content being summarized into the handoff document.
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute the shell command
mktemp -t handoff-XXXXXX.mdto generate a file path. Whilemktempis a standard utility, this involves direct command invocation and subsequent file system interaction (reading and writing to the temporary directory).
Audit Metadata