weekly-review

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to execute local Python scripts (scripts/open_loop_scanner.py, scripts/weekly_review_gate.py, and scripts/commitment_auditor.py) to automate the review process. These scripts are implemented using only the Python standard library and do not perform network requests or modify system configuration.
  • [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by reading data from potentially untrusted files in the user's workspace.
  • Ingestion points: scripts/open_loop_scanner.py reads markdown files and source code from user-specified directories to extract unchecked checkboxes and TODO/FIXME markers.
  • Boundary markers: No explicit delimiters or system instructions are used to distinguish extracted text snippets from the agent's primary instructions.
  • Capability inventory: The agent can read files, execute the provided Python scripts, and interact with the user to manage the review checklist.
  • Sanitization: Extracted text snippets are presented to the agent without escaping or sanitization, allowing for potential instructions embedded in scanned content to be interpreted by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 11:27 PM
Security Audit — agent-trust-hub — weekly-review