Skill v1.0.1

version: "1.0.1" name: gitlab-jobs description: "Manages GitLab CI jobs. Use for listing, logs, canceling, retrying, or erasing jobs. Triggers: CI builds, job status."

Overview

Covers job execution and logs.

Available Tools

• get_project_jobs: Retrieve a list of jobs for a specific GitLab project, optionally filtered by scope or a single job by id.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (Optional[int]): Optional. - Job ID
• scope (Optional[str]): Optional. - Filter jobs by scope (e.g., 'success', 'failed')
• include_retried (Optional[bool]): Optional. - Include retried jobs
• include_invisible (Optional[bool]): Optional. - Include invisible jobs (e.g., from hidden pipelines)
• get_project_job_log: Retrieve the log (trace) of a specific job in a GitLab project.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (int): Optional. - Job ID
• cancel_project_job: Cancel a specific job in a GitLab project.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (int): Optional. - Job ID
• ctx (Optional[Context]): Optional. - MCP context for progress
• retry_project_job: Retry a specific job in a GitLab project.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (int): Optional. - Job ID
• ctx (Optional[Context]): Optional. - MCP context for progress
• erase_project_job: Erase (delete artifacts and logs of) a specific job in a GitLab project.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (int): Optional. - Job ID
• ctx (Optional[Context]): Optional. - MCP context for progress
• run_project_job: Run (play) a specific manual job in a GitLab project.
• Parameters:
• project_id (str): Optional. - Project ID or path
• job_id (int): Optional. - Job ID
• ctx (Optional[Context]): Optional. - MCP context for progress
• get_pipeline_jobs: Retrieve a list of jobs for a specific pipeline in a GitLab project, optionally filtered by scope.
• Parameters:
• project_id (str): Optional. - Project ID or path
• pipeline_id (int): Optional. - Pipeline ID
• scope (Optional[str]): Optional. - Filter jobs by scope (e.g., 'success', 'failed')

Usage Instructions

1. Use job_id for actions.
2. Filters: scope, status.

Examples

• Get log: get_project_job_log with project_id="123", job_id=456.
• Retry: retry_project_job with project_id="123", job_id=456.

Error Handling

• Job not found: Verify IDs.