发现 Skills

Run code linting and static analysis. Use to check code quality, find issues, or auto-fix style problems.

Run a single PLAN.md inside a phase, performing atomic commits per task with deviation handling. Use this when a single plan inside a multi-plan phase is ready to be applied.

Develops resources for FiveM using the Qbox Project (qbx_core). Covers the exports-based API, bridge compatibility, Ox integration (ox_lib, ox_inventory), and best practices. Use when the user works with FiveM, Qbox, qbx_core, or mentions `exports.qbx_core`, `QBX.PlayerData`, or `ox_lib`.

Apply context-driven testing principles where practices are chosen based on project context, not universal 'best practices'. Use when making testing decisions, questioning dogma, or adapting approaches to specific project needs.

Measure quality effectively with actionable metrics. Use when establishing quality dashboards, defining KPIs, or evaluating test effectiveness.

AQE skill

A software security skill that integrates with Project CodeGuard to help AI coding agents write secure code and prevent common vulnerabilities. Use this skill when writing, reviewing, or modifying code to ensure secure-by-default practices are followed.

Investigate iOS CI builds - failures, performance, flaky tests, build times. Buildkite MCP, mobuild/Runway.

Expert guidance for creating effective Cursor slash commands with best practices, format requirements, and schema validation

Calculates pathway activity scores for metabolic pathways across different cell groups and subsets. This process quantifies the metabolic activity of each pathway per group, generating visualizations (heatmaps and violin plots) to compare metabolic states between clusters or conditions. Based on the methodology from Xiao et al.

Perform structured research and turn findings into an implementation plan.

Clojure-specific philosophy, idioms, and judgment frameworks. Use when working with Clojure code. Emphasizes data-oriented design, runtime validation with spec, REPL-driven development, and the philosophy of simplicity over ease.

Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution

Use this skill when working with Mass Entity, MassEntity, Mass AI, MassProcessor, MassFragment, MassTag, MassObserver, MassSpawner, MassCrowd, Mass ECS, entity archetype, ForEachEntityChunk, FMassEntityQuery, FMassEntityManager, ISM crowd, or large-scale entity simulation in Unreal Engine. See references/mass-entity-patterns.md for processor and observer templates. See references/mass-fragment-reference.md for built-in fragment types.

Use when the user is working with material, shader, MID, dynamic material, material instance, post-process, render target, parameter collection, decal, Nanite, Lumen, or rendering in Unreal Engine. See references/material-parameter-reference.md for parameter patterns and references/post-process-settings.md for post-process settings. For particle rendering, see ue-niagara-effects.

Administration reseau : diagnostic, DNS, firewall, ports, routing, VLAN, VPN, proxy, connectivite inter-VM.

This skill should be used when the user says "create a calendar event", "schedule a meeting", "add an event to my calendar", "book a time slot", "set up a meeting with", "add a Google Meet to an event", or wants to create a new Google Calendar event. Covers title, start/end time, attendees, location, description, and Meet link.

Use when main results pass result-to-claim (`claim_supported = yes` or `partial`) and ablation studies are needed for paper submission. A secondary Codex agent designs ablations from a reviewer's perspective; the local executor reviews feasibility and implements them.

Autonomous multi-round research review loop using MiniMax API. Use when you want to use MiniMax instead of Codex MCP for external review. Trigger with "auto review loop minimax" or "minimax review".

Autonomous design space exploration loop for computer architecture and EDA. Runs a program, analyzes results, tunes parameters, and iterates until objective is met or timeout. Use when user says "DSE", "design space exploration", "sweep parameters", "optimize", "find best config", or wants iterative parameter tuning.

Write detailed embodiment descriptions for patent specifications. Use when the user says 'draft embodiments', 'write embodiment', 'embodiment description', 'detailed description', or wants to describe how to practice an invention.

Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including General / Young / Outstanding Young / Distinguished Young Scholars / Overseas Outstanding Young / Key Program), ERC (EU), DFG (Germany), SNSF (Switzerland), ARC (Australia), NWO (Netherlands), and generic formats. Use when the user says "write grant", "grant proposal", "application", "write KAKENHI", "KAKENHI (科研費)", "funding application", "write a grant", "NSF proposal", or wants to turn research ideas into a funding application.

Workflow 1: Full idea discovery pipeline. Orchestrates /research-lit → /idea-creator → /novelty-check → /research-review to move from a broad research direction to validated, pilot-tested ideas. Use when the user says "full idea discovery process", "idea discovery pipeline", "start from scratch to find direction", or requests the complete idea exploration workflow.

Install and configure Python and uv. Use when setting up a Python environment.