智能体技能是一个包含指令、脚本和资源的文件夹，智能体按需加载。与 MCP 服务器不同，它不运行自己的进程，因此在上下文上更省，也更容易审查。
第 289–312 条，共 935 条
vaisala-xweather
This skill should be used when working with the Xweather MapsGL JavaScript SDK for the web (@xweather/mapsgl) — setting up a MapsGL map controller for Mapbox GL, MapLibre GL, Google Maps, or Leaflet, and adding, removing, styling, filtering, masking, or animating MapsGL weather layers and custom data layers. Use it whenever a task mentions MapsGL, aerisweather.mapsgl, addWeatherLayer, weather map
vaisala-xweather
This skill should be used to build Xweather Raster Maps image URLs (maps.api.xweather.com) — either static map images or XYZ map tile URLs for Leaflet, Mapbox, Google Maps, OpenLayers and similar libraries — from a description of the weather imagery wanted. Use it whenever a task mentions Raster Maps, maps.api.xweather.com, maps.aerisapi.com, an Xweather weather map layer or overlay (radar, satell
vaisala-xweather
Build and run Xweather Weather API request URLs for data.api.xweather.com from plain-language requirements. Use when a task mentions the Xweather or legacy Aeris API, weather endpoints such as observations, conditions, forecasts, alerts, lightning, air quality, tropical cyclones, tides, or road weather; asks for an API URL or query; needs help debugging an empty or failed request; asks about acces
smykla-skalski
List, reply to, resolve, and create GitHub PR review comment threads using gh CLI scripts. Use when managing code review feedback, replying to reviewer remarks, resolving review conversations, creating reviews with line-level comments, or bulk-processing threads by author.
smykla-skalski
Identify and remove AI writing patterns to make text sound natural and human-written. Use when humanizing commit messages, PR descriptions, review comments, docs, changelogs, or release notes. Also for de-slopping text that sounds robotic, has AI vibes, or reads like ChatGPT output.
smykla-skalski
Critique a Claude implementation plan before approving execution. Spawns parallel persona subagents (Skeptic, Architect, Verifier) to evaluate the plan against the codebase and return an Approve/Reject/Refine verdict with concrete refinements. Use when a plan has been generated (Plan Mode output, ExitPlanMode, pasted plan text, or plan file path) and needs review before code is written. Triggers o
smykla-skalski
Build and refine staff-level engineering resumes through interactive coaching, research-backed best practices, and per-job tailoring. Use when building, improving, or tailoring a resume for Staff/Principal engineer roles.
smykla-skalski
Clean up local branches with deleted remote tracking and their worktrees. Use after merging PRs to remove stale branches, detect squash-merged and rebased branches, and clean up associated worktrees.
diegocconsolini
Use this skill to reduce token consumption by 70-80% when using the Chrome DevTools MCP. Applies a snapshot-first strategy (text over images) with decision trees and pattern guides for navigation, forms, debugging, and visual checks, plus optional Gemini Flash vision processing for the cases that genuinely need an image.
smykla-skalski
Write tests that verify behavior (not implementation), use table-driven/parameterized patterns, and minimize mocking. Triggers when asked to write tests, add test coverage, or create test files. Also triggers when reviewing existing tests for quality.
vaisala-xweather
This skill should be used to design, build, secure, or debug an Xweather Webhooks receiver — the push alternative to polling the Weather API. Use it whenever a task mentions Xweather webhooks, pushed weather data, a weather webhook receiver or endpoint, subscribing to pushed hail/lightning/alerts/storm-cell data, or asks how to stop polling the Xweather API and receive data in real time instead. A
quay
Generic CVE triage: fetch advisories (NVD, GHSA, Go vuln DB), check package presence in dependency manifests, classify verdict, write assessment artifact, and post Jira comment. Project workflows supply repo mapping and branch policy.
smykla-skalski
Auto-review Go code for 100+ common mistakes when analyzing .go files, discussing Go patterns, or reviewing PRs with Go code. Checks error handling, concurrency, interfaces, performance, testing, and stdlib usage.
diegocconsolini
Use this skill to seed Claude Code's auto-memory with pre-built security/compliance patterns — known-safe code patterns, false-positive suppressions for the plugin-security-checker, org compliance rules, and supply-chain best practices. Helps Claude apply consistent security judgment across sessions.
quay
Ralph Loop tick-loop for single-ticket development. Replaces the /start -> /code -> /pr -> /poll skill chain with one continuous state machine. Each tick: read state, do one thing, write state, continue.
smykla-skalski
Refactoring review through seven sourced refactoring personas (Fowler, Uncle Bob, Feathers, Beck, Metz, Ousterhout, Tornhill) plus an adversarial pass, by reusing the source workflow under `claude/refactor-council`. Use when the user wants a refactoring review of code, a module, an app, or a diff.
everdict
Driving an evolution campaign — freeze a frame, log rounds whose verdict the platform derives, spend the adoption the gate authorizes. Use when running or designing a campaign, when a round comes back not-comparable, or when an adoption is refused.
diegocconsolini
Use this skill to extract and analyze large Word documents (1MB-50MB+) with minimal token usage. Losslessly extracts text, tables, formatting, and structure while achieving 10-50x token reduction through local extraction, semantic chunking by headings, and intelligent caching. Use when a .docx is too large for the context window (policies, contracts, technical reports).
diegocconsolini
Use this skill to extract and analyze large Excel workbooks (1MB-50MB+) with minimal token usage. Losslessly extracts formulas, cell formatting, and complex tables while achieving 20-100x token reduction through local extraction, semantic chunking, and intelligent caching. Use when an .xlsx is too large for the context window (compliance matrices, financial models, audit logs).
fworks-tech
Holds institutional knowledge about members, conventions, and architecture. Use before authoring new members or researching patterns.
mikonos
GTD skill 场景命令 · 搭建/自检可信系统。幂等建 memory/gtd/ 核心八清单 + 产品想法扩展清单 + 适配层自检 + 自动节律只读检查/显式安装 + 可选导入旧 open loops。首次运行入口。
everest3d
Ground any non-trivial C++ task in the bundled cppreference language and standard-library corpus, then verify target-specific details against current primary documentation. Use for learning, explaining, implementing, reviewing, debugging, fixing, porting, or optimizing C++ across hosted and embedded environments. Do not use it for pure C tasks unless C++ interoperability is part of the request.
smykla-skalski
Produce a daily AI news digest and reuse the source workflow under `claude/ai-daily-digest`. Use when the user wants a curated AI briefing or roundup in Codex.
smykla-skalski
Critique implementation plans by reusing the source workflow under `claude/plan-critic`. Use when the user wants a plan stress-tested before implementation.
