The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/check.py utilizes subprocess.run to execute internal validation tools such as validate_skill.py and package.py. These tools are used for linting and verifying the skill's structure and portability during development.
[EXTERNAL_DOWNLOADS]: The documentation includes instructions for installing the skill from its upstream source using npx, a standard package runner for Node.js.
[DATA_EXFILTRATION]: No exfiltration patterns were found. The validation scripts in scripts/asset_toolkit/ contain regular expressions to detect absolute system paths (e.g., /etc/, /root/). This functionality is used to audit the skill's own content for portability issues and does not represent an attempt to access or exfiltrate sensitive data.
[PROMPT_INJECTION]: The skill provides operational guidance and coordination logic for multi-agent workflows. It does not contain instructions intended to bypass safety guardrails, override system constraints, or extract internal prompts.

opencode-ensemble