claude-handoff
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to generate and run a shell command (claude --bg) using a summary of the current conversation. The summary, which includes untrusted user input and context, is interpolated directly into the command string. Without instructions for escaping shell metacharacters or quotes, this pattern creates a risk of command injection if the conversation contains characters like double quotes, backticks, or dollar signs.
- [PROMPT_INJECTION]: The skill implements a handoff mechanism that seeds a new background agent with data from the current session. Ingestion points: current conversation history and user-provided arguments in SKILL.md. Boundary markers: None present. Capability inventory: Shell command execution via 'claude --bg' in SKILL.md. Sanitization: No escaping or filtering of shell metacharacters is specified, although redacting sensitive info like API keys is mentioned. The lack of delimiters allows potentially malicious content to influence downstream agent behavior.
Audit Metadata