The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external project data in formats such as CSV, Excel, and JSON, which can be used to deliver malicious instructions to the agent.
Ingestion points: Project data and parameter modifications provided by the user via file paths or direct input as described in instructions.md and SKILL.md.
Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions to isolate user-provided data from the agent's core task instructions.
Capability inventory: The skill requires filesystem permissions for reading input files and exporting reports. The Python code performs numerical simulations using pandas and numpy but does not include network or system-level execution capabilities.
Sanitization: The instructions.md file provides some mitigation by requiring the agent to validate all inputs before processing and to report errors.

scenario-planner