claude-handoff
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the platform's native
claudeCLI to spawn background processes (claude --bg).- [INDIRECT_PROMPT_INJECTION]: The skill seeds new agent sessions with data derived from the current conversation and user arguments. - Ingestion points: Conversation history and user-supplied handoff arguments (SKILL.md).
- Boundary markers: None explicitly defined for the handoff prompt.
- Capability inventory: Execution of the
claudeCLI tool (SKILL.md). - Sanitization: The skill contains explicit instructions to redact sensitive information (API keys, passwords, PII) before creating the handoff summary.- [SAFE]: The storage of the handoff summary in a temporary directory is a standard mechanism for passing state between local processes. The skill's instructions for using the CLI tool specifically account for preventing unwanted shell expansion of the summary content.
Audit Metadata