ar-ap-aging-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface where scripts/aging_analyzer.py reads external financial data from a JSON configuration file. Ingestion points: scripts/aging_analyzer.py reads the config file via json.load. Boundary markers: Standard JSON structure is used, though specific instructions to ignore embedded prompts are absent. Capability inventory: The script is restricted to basic numerical analysis and terminal output; it contains no file system write operations, network requests, or subprocess execution capabilities. Sanitization: Structural validation is provided by the JSON parser. Due to the lack of exploitable capabilities, this ingestion surface is considered safe.
  • [COMMAND_EXECUTION]: The skill executes the local script scripts/aging_analyzer.py using the Python interpreter. The execution is limited to the bundled script and does not interpolate user-controlled strings into shell commands or sensitive system calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 12:23 AM