skills/aegntic/cldcde/debt-sentinel/Gen Agent Trust Hub

debt-sentinel

Warn

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute an external Python script (/a0/usr/plugins/debt-sentinel/debt-sentinel.py) via the Bash tool whenever code edits occur. This script is not provided within the skill package, preventing verification of its behavior.
  • [COMMAND_EXECUTION]: The troubleshooting documentation suggests running chmod +x on the external script path, which is a privilege-related command used to grant execution permissions to files.
  • [NO_CODE]: The core logic, including the enforcement script and its configuration files (ANTI_PATTERNS.md, DEBT.md), is missing from the skill distribution, making it an empty shell that relies on pre-existing local files.
  • [DATA_EXPOSURE_SURFACE]: The skill is designed to scan all file edits (Edit(*) or Write(*)) and log violations to a local file (DEBT.md). This broad access to the file system creates a surface where sensitive data encountered during edits could be unintentionally logged or handled by the unverified script.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 26, 2026, 08:45 PM