save
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bash command
kill -INT $CLAUDE_WRAPPER_PIDto terminate the process. While this is the intended functionality to exit the session, it involves direct shell command execution based on an environment variable. - [INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge between sessions by summarizing untrusted user data from the current conversation and writing it to a file (
tmp/handoff.md). - Ingestion points: Conversation history (SKILL.md, Step 1).
- Boundary markers: None; the skill does not specify delimiters or instructions for the agent to ignore potentially malicious content within the summarized text.
- Capability inventory: The skill has the ability to write files (
tmp/handoff.md) and execute shell commands (/hub next,kill). - Sanitization: There is no evidence of sanitization, escaping, or filtering applied to the conversation content before it is summarized and stored.
Audit Metadata