Agents, skills and MCP radar: skills and extensions for coding agents

Momentum based on available 7-day snapshots for Agents, skills and MCP: measured with two snapshots, estimated otherwise. Each value shows its own source window. This page reads stored snapshots only.

Radar filters

Tool ranking

Mixed ranking: measured growth takes priority; entries without two snapshots are estimated. Raw values keep their source-specific window, shown under each tool; a change requires at least two snapshots from the last 7 days.

  1. 1
    Active

    把 Markdown 一键排成可直接粘进公众号编辑器的精致 HTML —— 6 套精选主题 + 主题生成器 + 双关卡校验。An AI-agent skill that turns Markdown into paste-ready WeChat article HTML.

    githubmeasured growthOpen source ↗

    Install git clone https://github.com/isjiamu/gzh-design-skill ~/.claude/skills/gzh-design-skill

    3 536GitHub stars+104 (+3.0 %)
  2. 2
    Active

    Build tested agent skills and govern their lifecycle through a user-defined marketplace: evidence, discovery, updates, rollback, quarantine, and 17-platform distribution.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/FrancyJGLisboa/agent-skill-creator ~/.claude/skills/agent-skill-creator

    2 367GitHub stars·
  3. 3
    Active

    DeepSeek V4 × J-Space capability realization report — benchmark evidence that J-Space reduces capability-realization loss on DeepSeek V4.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Tiger3807861189/DeepSeek-V4-J-Space-Capability-Realization-Report ~/.claude/skills/DeepSeek-V4-J-Space-Capability-Realization-Report

    1 032GitHub stars·
  4. 4

    Skill
    Active

    Offline and security-first tool for syncing and managing agent skills

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/asteroid-belt/skulto ~/.claude/skills/skulto

    50GitHub stars·
  5. 5
    Active

    Agent skill for better research loops

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/nik1t7n/research-craft-skill ~/.claude/skills/research-craft-skill

    50GitHub stars·
  6. 6
    Active

    A LangGraph-powered multi-agent deep research system featuring task planning, human-in-the-loop review, multi-source retrieval, evidence-grounded report generation, full tracing and reproducible evaluation.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/hwfengcs/SDYJ_Multi_Agents ~/.claude/skills/SDYJ_Multi_Agents

    50GitHub stars·
  7. 7

    Skill
    Dormant

    A compact operating covenant for AI agents.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/chojondocho/vowline ~/.claude/skills/vowline

    50GitHub stars·
  8. 8

    Skill
    Active

    A light-weight and straightforward system for spec-driven development with Claude Code or OpenAI Codex. Written in Rust 🦀

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/marconae/speq-skill ~/.claude/skills/speq-skill

    50GitHub stars·
  9. 9
    Active

    Inspect, build, or extend Figma design systems from Claude Code. Read-only audits with WCAG checks and component scoring, build mode with variable bindings and slots, optional Phase 6 sync to tokens.css + AI rules.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/natdexterra/work-with-design-systems ~/.claude/skills/work-with-design-systems

    50GitHub stars·
  10. 10

    Skill
    Active

    205 production GTM agent skills for Claude Code — sales, outbound, prospecting, RevOps, ABM, PLG, CS, automation. Framework-cited playbooks with artifacts + QA scripts.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add LeadMagic/gtm-skills

    46GitHub stars·
  11. 11

    Skill
    Active

    Claude Desktop 的一站式中文补丁、字体补丁、会话管理、提示词研究

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add YX-hueimie/claude-omni

    13GitHub stars·
  12. 12
    Active

    Fan out heavy-lifting tasks from Claude Code (or other AI coding agents) to parallel Gemini sessions via the Antigravity CLI, then integrate the results.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add Vallykrie/gemini-swarm-skill

    6GitHub stars·
  13. 13

    Skill
    Dormant

    Structured, phase-gated architecture design powered by Claude Code. Enforces decision gates, challenges assumptions, and produces documented architecture decisions.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add AhmedHabiba/architor

    6GitHub stars·
  14. 14

    Skill
    Active

    sonmat (손맛) — Claude Code plugin that builds verification habits into AI-human collaboration

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add jun0-ds/sonmat

    6GitHub stars·
  15. 15

    Skill
    Dormant

    Turns your coding agent into a loud vulgar gym-bro chad.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add Stronger197/douchebag

    6GitHub stars·
  16. 16
    Dormant

    🎮 Arcade games for Claude Code — auto-playing pixel games driven by your live session's tokens, prompts, tool calls & context pressure. Single-plugin marketplace. Currently ships 🍄 Super Mario Runner.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add jaipatel248/claude-arcade

    6GitHub stars·
  17. 17

    Skill
    Active

    Claude Code skill: Chinese video model generation — Wan, PixVerse, Kling, Vidu, HappyHorse, Jimeng, MiniMax, Hunyuan. Text-to-video, image-to-video, first+last frame, reference-to-video.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Agents365-ai/videogencn ~/.claude/skills/videogencn

    4GitHub stars·
  18. 18

    Skill
    Dormant

    Seurat v5 single-cell analysis skill for Claude Code, OpenClaw, and Codex

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Agents365-ai/seurat-skill ~/.claude/skills/seurat-skill

    4GitHub stars·
  19. 19
    Active

    Lookup journal/magazine name abbreviation lookup (ISO 4 / MEDLINE)

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Agents365-ai/journal-abbrev ~/.claude/skills/journal-abbrev

    3GitHub stars·
  20. 20
    Active

    Multi-source drug-target due-diligence: ranked gene list → per-gene dossier across UniProt, OpenTargets, PubMed + cross-lineage DE scan + composite re-ranking. Claude Code / OpenClaw / SkillsMP skill.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Agents365-ai/target-prioritization ~/.claude/skills/target-prioritization

    3GitHub stars·
  21. 21

    Skill
    Active

    Multi-platform Chinese OCR skill — PaddleOCR/Baidu/Tencent/Alibaba/EasyOCR, all work in China

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/Agents365-ai/ocrCN ~/.claude/skills/ocrCN

    2GitHub stars·
  22. 22

    Skill
    Active

    Method and tools for systematic work with an AI development assistant — curated skills, subagents, CLI, and a working methodology, plus a VS Code setup for macOS.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/KirKruglov/OFFICINA ~/.claude/skills/OFFICINA

    1GitHub stars·
  23. 23
    Active

    Evidence-first, Fact/Evidence/Assumption/Unknown teardown of a startup or app idea — solo-build screen or VC-scale investment teardown

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/sakshi-gupta-1/idea-teardown-skill ~/.claude/skills/idea-teardown-skill

    1GitHub stars·
  24. 24

    Skill
    Active

    Catches the exact security mistakes that have actually breached real AI-built apps, before you deploy.

    githubestimated momentumOpen source ↗

    Install git clone https://github.com/AmirHosein-Lotfi/Seaworthy ~/.claude/skills/Seaworthy

    1GitHub stars·

Learning resources

Estimated ranking: fewer than two daily snapshots are not enough to measure growth. These resources remain available separately and do not take part in the main tool ranking.

  1. 1

    tunelab

    Skill
    ActiveOpen source ↗

    Claude Code plugin for LLM fine-tuning, distillation, and evaluation — decide whether you need fine-tuning at all, distill your LLM logs into small local models (MLX/LoRA), evaluate with held-out discipline, and learn the why at every step.

    github

    Install /plugin marketplace add rchaz/tunelab

    Resourceestimated momentum
    6GitHub stars·