The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script scripts/validate_map.py to perform structural validation on JSON maps. This is a benign use of local execution for a defined development task.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided JSON maps, which constitutes an ingestion surface. 1. Ingestion points: Data enters via the JSON map file passed to the validation script. 2. Boundary markers: None explicitly defined in the instructions. 3. Capability inventory: The skill leverages python3 for script execution but lacks network or sensitive data access. 4. Sanitization: The validation script applies strict schema and type checks to the input data.

narrative-scrollytelling