工具说明为英文。
技术工具雷达: Agents, skills and MCP
雷达筛选条件
工具排名
按实测增长排序，并在不同来源间归一化。 原始值保留各自的数据窗口；计算变化需要 7 天内至少两次测量。
- 1活跃
taste-skill
智能体Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop
github实测增长打开来源 ↗
82 205GitHub 星标+643 (+0.79 %) - 2活跃
Build production marketing sites with Claude Code — an Astro + Cloudflare Workers template, a WordPress→Astro migration playbook, and 36 silent failures documented from real rebuilds.
github实测增长打开来源 ↗
21GitHub 星标+10 (+90.9 %) - 3活跃
COSCUP 2026 demo: Microsoft Agent Framework composes a remote LangGraph agent over A2A v1, with Microsoft Foundry live and credential-free offline paths.
github实测增长打开来源 ↗
4GitHub 星标+4 - 4活跃
Extracted system prompts from Anthropic - Claude Fable 5, Opus 5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-5.6-Sol, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more.…
github实测增长打开来源 ↗
63 972GitHub 星标+206 (+0.32 %) - 5活跃
a2acode
智能体Serve any ACP coding agent — Claude Code, Gemini CLI, Codex — over the A2A protocol, with tool progress, file diffs, and permission prompts preserved.
github实测增长打开来源 ↗
4GitHub 星标+2 (+100.0 %) - 6活跃
a2acast
智能体Zero-infrastructure A2A messaging between AI agents on different machines — Claude, ChatGPT/Codex, Copilot, any harness; ntfy relay transport, no ports, no server
github实测增长打开来源 ↗
8GitHub 星标+3 (+60.0 %) - 79GitHub 星标+3 (+50.0 %)
- 8活跃
openamer
智能体The self-improving, self-learning open-source AI agent. Background Computer-Use, A2A Swarm, Vector Memory, Plugin System, Voice, Smart Approvals - 100/100 Score. Windows-native.
github实测增长打开来源 ↗
2GitHub 星标+1 (+100.0 %) - 9活跃
intentfence
智能体Payment preflight and x402-paid signed audit receipts for autonomous AI agents. REST, MCP, A2A, Base USDC.
github实测增长打开来源 ↗
2GitHub 星标+1 (+100.0 %) - 10活跃
registry
智能体A registry website for hosting the a2a agents that were built using the Inference Gateway ADK
github实测增长打开来源 ↗
2GitHub 星标+1 (+100.0 %) - 11活跃
Autonomous Execution Gateway: before AI agents execute tools, payments, or workflows, InterAI verifies risk, policy, and trust receipts.
github实测增长打开来源 ↗
2GitHub 星标+1 (+100.0 %) - 12活跃
historian
智能体A local A2A event and continuity service for agent applications, with structured history, provenance-safe transcripts, literal search, and natural-language queries.
github实测增长打开来源 ↗
1GitHub 星标+1 - 13活跃
legatio-ai
智能体Agent Authorization & Negotiation Gateway. A policy engine that lets AI agents act on behalf of humans with enforceable rules, human approval, and audit trails.
github实测增长打开来源 ↗
1GitHub 星标+1 - 14活跃
Attenu Guard for TypeScript — per-agent permissions on every tool call and sub-agent handoff (a sub-agent never holds more than its parent) with a hash-chained audit log verified offline; same bundle format as the Python library.
github实测增长打开来源 ↗
1GitHub 星标+1 - 1510GitHub 星标+3 (+42.9 %)
- 16活跃
Multi-agent orchestration for DeepSeek Harness: turn isolated sessions into a team that assigns work, reviews output, and runs DAG workflows — durable task ledger, real reviewers, crash-safe recovery.
github实测增长打开来源 ↗
6GitHub 星标+2 (+50.0 %) - 17活跃
review-squad
智能体Dispatch panels of specialized subagents to review projects — expert audits, first-impression personas, task-completion flows, and pedantic nitpicks
github实测增长打开来源 ↗
12GitHub 星标+3 (+33.3 %) - 183GitHub 星标+1 (+50.0 %)
- 19休眠
meepo
智能体A local, privacy-first AI agent (Rust) that connects LLMs to your email, calendar, browser, code, and messages — 75+ tools, MCP/A2A support, Discord/Slack/iMessage channels, autonomous agent loop.
github实测增长打开来源 ↗
3GitHub 星标+1 (+50.0 %) - 20活跃
investagent
智能体全栈AI投研Agent · Your AI research co-pilot for smarter investment decisions. Five integrated research frameworks: Serenity industry-chain analysis, Buffett value-investing filter, UZI-Skill 22-dim deep analysis, TradingAgents multi-agent…
github实测增长打开来源 ↗
10GitHub 星标+2 (+25.0 %) - 21休眠
HERALD
智能体HERALD is an open-source CLI that emits the standard discovery files agents need to read your site and discover its agent-interaction capabilitie. One config object drives all of it. Each file is an independent open standard; pick the…
github实测增长打开来源 ↗
4GitHub 星标+1 (+33.3 %) - 22活跃
한국 취준생 자소서 작성 멀티 에이전트 Claude Code 스킬 심층 인터뷰로 소재 발굴, 경험카드·claim-map으로 날조 차단, 사용자 문체 반영
github实测增长打开来源 ↗
4GitHub 星标+1 (+33.3 %) - 23活跃
council
智能体고개만 끄덕이는 자문단은 잊으세요. 25인 거장 자문단이 당신의 어려운 결정을 로스팅합니다. AI ROASTING 자문단 · Claude Code 스킬
github实测增长打开来源 ↗
4GitHub 星标+1 (+33.3 %)
学习与参考资源
按实测增长排序，并在不同来源间归一化。 这些资源可单独访问，不参与主要排名。
- 1活跃打开来源 ↗
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
github资源实测增长75 687GitHub 星标+196 (+0.26 %) - 2活跃打开来源 ↗
30x-outreach
智能体Auditable outbound experiments from verified signals to deterministic quality gates, exact-payload approval, safe execution, and compounding learning.
github资源实测增长3GitHub 星标+1 (+50.0 %)