:)
zereight publishes agent skills that cover the working loop of software development: GitLab merge request, issue and pipeline work through the GitLab MCP server, security sweeps for code changes and dependencies, test-driven development, cleanup of AI-generated code, and cross-language coding standards. Their most popular skills also include consensus planning with an iterative review loop and status reporting for active agent modes, with a practical focus on how agents plan, verify and hand off work.
Gitlab Mcp
24 skills · 2,089 stars · MIT
Ai Slop Cleaner Clean AI-generated code slop with regression-safe workflow.
Cancel Cancel any active OMG mode and clean up state.
Ccg Claude-Codex-Gemini tri-model orchestration for multi-perspective analysis.
Coding Standards Canonical cross-language coding standards reference.
Deep Dive 2-stage pipeline: trace (causal investigation) then deep-interview (requirements).
Deep Interview Socratic deep interview with mathematical ambiguity gating.
Gitlab Mcp Skill Use this skill when working with the GitLab MCP server tools for merge requests, issues, reposito…
Gl Cli Use GitLab from the terminal without registering an MCP server.
Omg Autopilot Full autonomous execution from idea to working code.
Plan Strategic planning with optional interview workflow.
Ralph PRD-driven persistence loop until task completion with verification.
Ralplan Consensus planning with Planner/Architect/Critic iterative loop.
Remember Promote reusable project knowledge to the right memory surface.
Review Code review with spec compliance and quality checks.
Security Scan Rapid security scanning workflow for code changes.
Self Improve Autonomous evolutionary code improvement engine with tournament selection.
Skill Stocktake Audit the skill inventory for quality, coverage, and staleness.
Status Show current OMG mode status, active agents, and progress.
Tdd Test-Driven Development enforcement skill.
Team N coordinated agents on shared task list with staged pipeline.
Trace Evidence-driven tracing with competing hypotheses in parallel.
Ultraqa QA cycling workflow - test, verify, fix, repeat until goal met.
Ultrawork Parallel execution engine for high-throughput task completion.
Verify Verify that a change really works before claiming completion.