strategic-compact
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation instructs the user to configure a PreToolUse hook that executes a local Node.js script (suggest-compact.js). This script is automatically invoked whenever the agent attempts to use the Edit or Write tools.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process session transcripts, which include untrusted content from the conversation history and tool outputs.
- Ingestion points: The session transcript is accessed via the transcript_path provided in the hook payload.
- Boundary markers: The documentation does not describe any specific delimiters or instructions used to isolate the processed transcript data from the script's execution context.
- Capability inventory: The skill triggers the execution of a shell command (node) via the PreToolUse hook whenever core agent editing tools are invoked.
- Sanitization: There is no evidence of sanitization or validation of the transcript content before it is processed by the script.
- [NO_CODE]: The primary logic component, suggest-compact.js, is not included in the provided files, which prevents a full verification of the script's behavior and data handling.
Audit Metadata