The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the stable-baselines3 package and its dependencies using uv pip. This library is a well-known, community-standard tool for reinforcement learning development.\n- [PROMPT_INJECTION]: The skill facilitates the creation of custom environments that process external observation data, creating a surface for indirect prompt injection.\n
Ingestion points: Observation data is processed through the step and reset methods in scripts/custom_env_template.py and described in references/custom_environments.md.\n
Boundary markers: The provided templates do not implement boundary markers or instructions to ignore embedded commands within observation data.\n
Capability inventory: The skill's scripts possess capabilities for file system modification (model.save in scripts/train_rl_agent.py) and process management (SubprocVecEnv in scripts/train_rl_agent.py).\n
Sanitization: No explicit sanitization or input validation for environment observations is provided in the boilerplate code.\n- [SAFE]: All core functionalities, including model saving, loading, and vectorized environment usage, follow established best practices for machine learning development and contain no malicious logic.

stable-baselines3