The Agent Skills Directory

[SAFE]: The skill provides instructions and references for managing lab machines (Opentrons, Balance, etc.). No malicious patterns were detected across the analyzed files.
[COMMAND_EXECUTION]: The skill instructions involve running local Python scripts using the uv package manager (uv run --package balance-edge python edge/balance.py) and calling a domain-specific CLI tool puda. These are standard operations for the skill's stated purpose of lab automation.
[EXTERNAL_DOWNLOADS]: The documentation mentions using uv sync --all-packages, which is a standard Python dependency management command. No suspicious remote script execution patterns (e.g., curl | bash) or downloads from untrusted sources were detected.
[PROMPT_INJECTION]: The skill includes robust instructions (labeled as "Critical Rules") directing the agent to ask the user for clarification rather than making assumptions about machine selection or configuration. This is a security best practice that reduces the risk of unintended actions.

bears-machines