weekly-review
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
scripts/open_loop_scanner.pyscript to scan user-defined directories (e.g.,~/notes) for checkboxes and TODO markers. If these files contain malicious instructions disguised as notes or tasks, the agent could be manipulated when it reviews the scanner's output.\n - Ingestion points: Text data is read from files within user-controlled directories by
scripts/open_loop_scanner.pyin the 'GET CLEAR' phase.\n - Boundary markers: The skill does not include specific delimiters or 'ignore' instructions to help the agent distinguish between its own operational output and the content found in user files.\n
- Capability inventory: The skill allows the agent to execute local Python scripts and interact with task lists, which could be exploited if the agent is successfully injected.\n
- Sanitization: No sanitization is performed on the extracted text snippets; they are only truncated for display.
Audit Metadata