The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using curl and jq to interact with the GitLab API. These operations are limited to standard API endpoints and use environment variables for configuration.
[EXTERNAL_DOWNLOADS]: Data is fetched from the GitLab API at the location specified by the user's GITLAB_URL environment variable. This is necessary for the skill's functionality and does not involve downloading executable scripts.
[CREDENTIALS_UNSAFE]: The skill uses the GITLAB_TOKEN environment variable for authentication. This is a secure practice as it avoids hardcoding sensitive credentials directly in the skill file.
[PROMPT_INJECTION]: The skill processes data from GitLab, such as merge request descriptions and job logs. This represents a surface for indirect prompt injection, but the skill itself does not contain instructions to bypass safety guidelines or override agent behavior.

gitlab