Инструменты

Connects the open-source ERPNext to AI assistants: query invoices, customers and inventory, run reports and handle approvals in plain language, with permissions following ERPNext accounts and every call audit-logged. An admin must install the app first; have finance check the analysis.

Vercel's official plugin: a knowledge graph of the Vercel ecosystem, dozens of skills, expert subagents for deployment and performance, and commands for deploys, env vars and project status, covering Next.js, the AI SDK, storage and CI/CD. Needs Node.js 18+ and Bun; telemetry is opt-out.

Todoist's official td CLI: quickly add tasks in natural language, view Today, Inbox and project to-dos, authorize read-only if you like, and install companion skills for Claude Code, Codex and others in one step. Needs a Todoist account.

CFO advisor skill from Every for bootstrapped small companies: helps founders model cash runway, burn rate and unit economics (LTV:CAC, payback), assess hiring ROI, and build 13-week cash flow and scenario forecasts. Framework-only, no account; check big decisions with a pro.

QuantConnect's official CLI drives the open-source LEAN engine locally or in the cloud to download data, backtest and optimize quant strategies and deploy them live. Requires a paid-tier QuantConnect organization; strategies in Python or C#; local runs need Docker. Get expert review.

Postman's official MCP server: lets agents manage workspaces, collections, environments and API specs, help with API testing and generate client code from API definitions. The remote version uses OAuth; the local version needs a Postman API key.

Official Readwise skills: triage your Reader queue, skim feeds, quiz you on what you just read, and draft book reviews and reading recaps from your highlights. Needs a Readwise subscription (30-day free trial) and the Readwise MCP or CLI installed first.

Hugging Face's official MCP server: lets agents search models, datasets and Spaces on the Hub and call thousands of Gradio AI apps directly. Hosted remotely; sign in with your Hugging Face account or authorize with a token.

Rewrites prose to remove 41 AI-writing patterns, from significance inflation and promotional tone to em dashes and chatbot artifacts, using a draft, audit, final loop that never invents facts. Can match a writing sample, edit files in place, or audit a whole docs folder for slop.

Lets agents search Wikipedia and read full articles, summaries, specific sections and related articles, useful for lesson prep, homework help and fact-checking. English Wikipedia by default; pass --language to use another edition. No account; needs uv (Python).

Detects 55 AI-writing patterns and rewrites text in one of five voices (casual, professional, technical, warm, blunt), with an optional 0-100 AI-tell score. Offers detect, rewrite and in-place Markdown edit modes and never invents facts. Pure Markdown, no dependencies or network calls.

Basecamp's official CLI: manage projects, to-dos, card tables, messages and files, post to project chat and search across projects, with JSON output with next-step hints, plus a bundled agent skill and Claude Code and Codex plugins. Needs a Basecamp account; the installer handles browser auth.

B2B prospecting plugin from Explorium: filter companies and contacts by industry, size and tech stack, enrich CRM leads, look up business events such as hiring and funding, and export CSV lists. Needs a Vibe Prospecting account and, in Claude Code, the vpai CLI (Node.js). Respect privacy laws.

Alibaba Cloud's official agent skills: ECS diagnosis and DR, RDS/PolarDB database checks, PAI training and EAS deployment, ACK clusters, RAM permissions, OSS and network troubleshooting, and Model Studio. Needs an AccessKey or aliyun CLI credentials.

Turns Markdown outlines into interactive mind map HTML and can export PNG, JPG or SVG server-side (requires Playwright). Runs fully locally with no third-party API key; good for organizing notes, lesson prep and document structure.

Connects to Actual Budget, the open-source budgeting app: query accounts and transactions, analyze spending by category and payee, compare budget to actual, track cash flow and net worth; --enable-write adds editing of transactions, categories and rules. Needs an Actual server or local data.

Firecrawl's official plugin: brings the Firecrawl CLI in as skills so Claude Code can search the web, scrape pages to Markdown, map site links or crawl whole sites. Install firecrawl-cli and set an API key first; self-hosting supported.

i18next's official CLI: finds hard-coded strings in JS/TS code, extracts translation keys, checks which locales are missing entries and syncs locale files; init --yes sets up non-interactively, and a bundled agent skill guides app internationalization. Needs a Node.js project using i18next.

Nulab's official Backlog MCP: lets agents manage Backlog projects, issues and comments, milestones and wikis, and view Git repositories and pull requests. Needs your Backlog space URL and a personal API key, plus Node.js or Docker.

Searches Unsplash's free high-res photos by keyword, color and orientation to find images for posters, articles, slides and designs. Needs a free Access Key from the Unsplash developer portal; clone the repo and run the Python server with uv.

Official MCP from Razorpay, the Indian payments platform: query payments, orders, refunds, settlements and reconciliation reports, and create payment links and QR codes. The hosted remote server is recommended; generate a merchant token from your API key and secret.

MinerU's official MCP: parses PDF, Word, PowerPoint and images into Markdown via the mineru.net cloud API, keeping tables and formulas. The keyless Flash mode works out of the box; add a token for higher quotas and more formats.

Comet's official Opik MCP: query traces and spans of LLM apps in chat, log scores and comments, and manage datasets, experiments and prompt versions. Works with Opik Cloud (API key or browser sign-in) and self-hosted open-source Opik.

Diagnoses page-load and interaction performance from Chrome DevTools traces, network evidence and source code, covering Core Web Vitals, LCP and layout shifts. Separates measured values from estimates and hypotheses, and re-measures fixes under equivalent conditions.