From gitlab-skills
List all open GitLab merge requests for the current project. Use when a user asks to show, list, or review open merge requests in the repository.
How this skill is triggered — by the user, by Claude, or both
Slash command
/gitlab-skills:gitlab-mr-list-openThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
List all open merge requests for the current GitLab project. The skill uses the
List all open merge requests for the current GitLab project. The skill uses the
GitLab CLI when GITLAB_TOKEN is not set and the REST API when the token is
available.
$ARGUMENTS is optional. Ignore extra arguments unless the user explicitly asks
for filtering that this skill supports in the future.
Check whether GITLAB_TOKEN is set.
Run:
glab mr list -F json -o updated_at -S desc
Display the command output as returned. If the command fails, show the CLI error output clearly.
Require these environment variables and fail clearly if any are missing:
GITLAB_HOSTGITLAB_TOKENDetermine the project path from git remote get-url origin, strip the host
prefix and .git suffix, then URL-encode / as %2F.
Fetch open merge requests for the current project:
GET {GITLAB_HOST}/api/v4/projects/{encoded_project_path}/merge_requests?state=opened&order_by=updated_at&sort=desc
Headers: PRIVATE-TOKEN: {GITLAB_TOKEN}
For each merge request, display:
If the response is empty, report that there are no open merge requests. If the API returns an error, report the HTTP status and response message.
Return a concise list ordered by most recently updated merge request first. Each entry should include the MR IID, title, author, branch mapping, and URL. On failure, include the HTTP status or CLI error output.
/gitlab-mr-list-open
list open merge requests
show open MRs for this project
GITLAB_TOKEN is available.read_api scope is sufficient for the API path.npx claudepluginhub rogerbriggen/ai_skills --plugin gitlab-skillsCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.