scenario-planning-engine
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's instructions and associated Python script do not contain any malicious patterns, prompt injection attempts, or unauthorized system access. The script performs localized mathematical calculations based on structured user input.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a user-supplied JSON file which is then processed and output to the agent's context. This represents a potential surface for indirect prompt injection if the source data contains malicious instructions. 1. Ingestion points: Data is read from a JSON file via the --config argument in scripts/scenario_engine.py. 2. Boundary markers: The script provides structured output but lacks explicit delimiters to prevent the agent from interpreting data as instructions. 3. Capability inventory: The script's operations are limited to reading local files, arithmetic calculations, and printing to standard output. 4. Sanitization: The script validates numerical probability sums but does not sanitize or escape string inputs like metric names before printing them.
Audit Metadata