ツール

Detects your CPU, memory and GPU and scores open models on whether they run, speed, quality and context, then works with local runtimes like Ollama, llama.cpp and MLX to download and benchmark them. recommend --json gives agent-friendly output.

Local model gateway that routes requests from Claude Code, Codex and other coding agents to providers like DeepSeek, Kimi, Gemini and OpenRouter, with failover, rule-based routing and request logs. Bring your own API keys.

Alibaba's open-source Java production diagnostics tool: inspect method arguments and timing, threads and GC, class loading and decompiled code, and generate flame graphs without changing code or restarting the JVM. Supports JDK 8+ via an interactive CLI.

Headless browser written from scratch in Zig for AI agents and automation, using less memory and running faster than headless Chrome in official benchmarks. fetch turns pages into Markdown, and it can serve CDP or MCP. Telemetry is on by default (an env var disables it); Windows needs WSL.

Trading research agent from HKU's HKUDS whose MCP server offers 70+ tools for quotes, financials, backtesting and research across A-shares, Hong Kong and US stocks and crypto. Needs local Python; HK, US and crypto research need no API key. For research only; not investment advice.

Package manager for the open Agent Skills ecosystem: one command installs skills from GitHub and other sources into Claude Code, Codex, Cursor and dozens more agents, with search, update and removal. From Vercel Labs.

When a user asks whether a skill exists for some task, searches the skills.sh ecosystem with the skills CLI, then recommends and installs a fit. Maintained by Vercel Labs and shipped in the skills CLI repo.

HashiCorp's secrets management system: the vault CLI reads and writes secrets, issues dynamic database and cloud credentials on demand, encrypts and decrypts data, and manages leases and revocation. Needs a reachable Vault server.

Command runner that saves and runs project commands in a justfile, with make-inspired but simpler syntax. Agents can use just --list to discover a project's build, test and deploy commands and run them directly.

Claude Code plugin for Next AI Draw.io that starts its Draw.io MCP server via npx. Generate and edit flowcharts, architecture and sequence diagrams in plain language, with live browser preview, version history and .drawio export. Needs Node.js.

Lets agents create and edit draw.io diagrams such as flowcharts and architecture diagrams in plain language, with live browser preview, version history and export to .drawio, PNG or SVG. Includes an embedded server; runs locally with no API key.

Lightweight command-line JSON processor that slices, filters, maps and transforms JSON with concise expressions and no runtime dependencies. A staple for agents parsing API responses and logs.

Terminal multiplexer written in Rust, an alternative to tmux, with layout files, floating and stacked panes, WebAssembly plugins and a built-in web client. Handy for running several agent sessions at once.

Official CLI for Tencent's open-source knowledge base platform WeKnora: upload documents, manage knowledge bases, run hybrid search and answer questions from them, or serve as a read-only MCP. Needs your own WeKnora deployment and a source build with Go 1.26+.

Polyglot dev environment manager: declare Node, Python, Go and other tool versions, environment variables and project tasks in mise.toml; mise exec and mise run install tools on demand and run them, with one config shared by local and CI.

Community-maintained cybersecurity skill library: 800+ skills across 34 domains including threat hunting, incident forensics, penetration testing and cloud security, mapped to MITRE ATT&CK and NIST CSF. Not by Anthropic; offensive skills are for authorized testing only.

Official Slack (Salesforce) plugin in Anthropic's knowledge work marketplace: connects Slack's hosted MCP to search and send messages, read channels and threads, and manage canvases, with commands for channel digests, standup updates and announcements. Needs workspace admin approval.

Triages support tickets: extracts the issue and urgency signals, categorizes and assigns P1 to P4 priority, checks for duplicates or known issues, suggests the team to route to and drafts the first response. Part of Anthropic's official customer-support plugin.

Drafts customer-facing replies tuned to the situation and relationship: product questions, outage notices, bad news like delays or won't-fix, declined feature requests and billing disputes, with adjustable tone. From Anthropic's official customer-support plugin; review before sending.

Support for small merchants: reads customer emails, checks orders and refunds in Stripe, PayPal, Square or Shopify, and drafts replies in the owner's voice per your refund policy; can refund via Stripe or Square after explicit approval. Works with pasted emails too. Official from Anthropic.

Packages customer issues that need engineering, product or leadership into an escalation brief with scope, repro steps, business impact, asks and deadline, and decides who to escalate to. Support and CRM connectors add context automatically. Part of Anthropic's official customer-support plugin.

Pulls verbatim customer quotes on a topic from call transcripts, meeting notes and emails, tagging each with speaker, account and date, and summarizes recurring themes into voice-of-customer findings for product and marketing. Connect Gong, email or drive, or upload transcripts. By Anthropic.

Account health checks for customer success managers: pulls CRM, email, calendar and call notes to score relationship, engagement, commercial (renewal and expansion), support and value delivery red/yellow/green, and can prep a QBR agenda. CRM preferred; uploads work. Official from Anthropic.

Collects reviews and complaints from Google, Yelp, Facebook and more, groups them by theme with quotes, and drafts replies in the owner's voice that post only after approval; also finds lapsed customers and drafts win-back messages. Official Anthropic small-business plugin.