frappe-agent-debugger

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes bench CLI commands for site administration, database inspection, and log monitoring.\n- [REMOTE_CODE_EXECUTION]: Includes documentation for remote debugging attachment with debugpy and dependency management using bench pip install.\n- [PROMPT_INJECTION]: Identified an indirect prompt injection surface when the agent processes logs or tracebacks. \n
  • Ingestion points: Application logs, system logs, and Error Log DocType entries.\n
  • Boundary markers: No explicit delimiters used for external data.\n
  • Capability inventory: Ability to run shell commands, perform SQL queries, and read system files.\n
  • Sanitization: Standard SQL parameterization is recommended, but no content sanitization for processed log entries is provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 11:52 AM