The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of local simulation tools (rocket-sim) and optimization scripts (optimizer.py) within the user's environment to process rocket design files.
[PROMPT_INJECTION]: The skill implements an AI Optimizer loop that processes .ork design files and simulation outputs to generate design iterations. This creates an indirect prompt injection surface where rocket designs or simulation data could attempt to influence the agent's logic. (1) Ingestion points: .ork files and simulation result JSON data processed in the optimizer.py loop. (2) Boundary markers: None identified; external data is used directly to guide the optimization process. (3) Capability inventory: Execution of shell commands and Python scripts via subprocesses to run simulations. (4) Sanitization: No explicit sanitization or validation of the simulation data or design files is described before processing by the LLM.
[EXTERNAL_DOWNLOADS]: The skill references the official OpenRocket project on GitHub for its simulation core. This is a well-known and legitimate resource in the aerospace community.

openrocket-sim