--- name: load-mr-context description: Load MR context for the current branch using GitLab MCP tools --- # Load MR Context ## Name gitlab:load-mr-context - Load merge request context for the current branch ## Synopsis ``` /load-mr-context [project-path] [branch] ``` ## Description Loads the full context of an open merge request for the given branch, including description, changes, pipeline status, and review comments. Uses GitLab MCP tools for structured data access. ## Implementation Given `project-path` and `branch` arguments: 1. Use `mcp__plugin_gitlab_gitlab__list_merge_requests` with `project_id: "PROJECT_PATH"`, `source_branch: "BRANCH"`, `state: "opened"` to find the MR 2. If no MR is found, report that no open MR exists for this branch and stop 3. If an MR exists, gather context **silently** (do not narrate each step): - Use `mcp__plugin_gitlab_gitlab__get_merge_request` with the MR IID to read description, labels, reviewers, and metadata - Use `mcp__plugin_gitlab_gitlab__get_merge_request_diffs` to understand what has changed - Check pipeline status for the branch - Check for review notes/discussions - Check for linked issues mentioned in the MR description 4. Present a concise summary: - MR title, number, and status - Key points from the description - Files changed (count and notable files) - Pipeline status (passing/failing) - Review status (approved, changes requested, pending) - Any blocking issues or failing pipelines ## Example Interaction ``` User: /load-mr-context my-group/my-project feat/mr-context Claude: ## MR !42: Add MR context detection hooks - **Status**: Open, 1 review pending - **Description**: Adds SessionStart hooks to GitLab plugin that detect open MRs on the current branch - **Changes**: 4 files (+120, -8) - **Pipeline**: All stages passing - **Reviews**: 1 approved, 1 pending from @reviewer Ready to assist with this MR. ``` ## Arguments - `project-path` (required): GitLab project path (e.g., `group/project` or numeric ID) - `branch` (required): Branch name to find the MR for ## Tips - This skill is typically invoked automatically when starting a session on a feature branch - All data is fetched via MCP tools (no CLI commands needed) - The summary is kept concise to avoid cluttering the session start