The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses eval to process shell-formatted output from the mergify ci git-refs command. This is a standard mechanism to export variables such as MERGIFY_GIT_REFS_BASE into the current shell session.
[DATA_EXFILTRATION]: The skill transmits test result metadata and JUnit XML content to https://api.mergify.com. This is the primary function of the tool and targets the official Mergify API.
[CREDENTIALS_UNSAFE]: Sensitive authentication is handled via the MERGIFY_TOKEN environment variable. The skill documentation explicitly recommends using GitHub Secrets or equivalent secure storage, which aligns with industry best practices.

mergify-ci