The Agent Skills Directory

[SAFE]: The skill implements building energy simulation logic using standard Python libraries such as numpy and dataclasses. All calculations are performed locally.
[INDIRECT_PROMPT_INJECTION]: The skill processes project data from external formats (CSV, Excel, JSON) which constitutes an ingestion surface for untrusted data.
Ingestion points: Data ingestion occurs through user-provided project parameters and files as described in instructions.md and SKILL.md.
Boundary markers: No specific boundary markers are defined in the instructions for separating data from instructions.
Capability inventory: The skill possesses filesystem access for reading and writing data, but does not use eval, exec, or network operations.
Sanitization: The instructions.md file explicitly mandates that the assistant must validate inputs before processing and follow industry standards.

energy-simulation