The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides detailed instructions for executing the runhuman CLI tool to manage jobs, projects, organizations, and templates. It also utilizes the gh (GitHub) CLI for repository and issue management.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch an external guide from https://runhuman.com/for_agents_github_actions.md and check for the latest releases of volter-ai/runhuman-action on GitHub. These operations target the official domains and repositories associated with the skill's author and product.
[CREDENTIALS_UNSAFE]: The skill correctly directs agents to handle authentication via the runhuman login flow or using environment variables like RUNHUMAN_API_KEY, avoiding hardcoded secrets and following standard security practices for CLI tools.
[DATA_EXFILTRATION]: While the skill interacts with a remote API (runhuman.com), all network traffic is associated with the core functionality of the service. No suspicious or unauthorized data exfiltration patterns were detected.

runhuman-testing