working-capital-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external configuration data and incorporating it into the agent's context.
- Ingestion points: The script
scripts/working_capital.pyreads a JSON configuration file containing user-defined descriptive labels for cash-release levers. - Boundary markers: The script output does not use specific delimiters or instructions to prevent the agent from interpreting content within the lever names as instructions.
- Capability inventory: The skill executes a local Python script and reads its stdout; it has broad read access to financial data provided in the config.
- Sanitization: There is no evidence of string sanitization or validation for the
namefield in the input configuration before it is printed to the output and read by the agent.
Audit Metadata