month-end-close-accelerator
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external configuration data which creates a potential surface for indirect prompt injection.
- Ingestion points: The
scripts/close_tracker.pyscript ingests data from a user-specified JSON file (passed via the--configargument). - Boundary markers: Absent. The script prints raw string values from the JSON (such as task names and account descriptions) to the console output.
- Capability inventory: Analysis of the provided scripts confirms that the tool is limited to basic file reading and terminal output. It does not possess capabilities for network communication, persistent storage, or arbitrary command execution.
- Sanitization: The script performs standard JSON parsing but does not sanitize or validate the content of text fields before they are outputted to the agent's context.
Audit Metadata