Best Focus AI Skills & MCP Servers

Semantic search using TF-IDF vector space — cosine similarity, intent matching, document similarity.

Task management for multi-agent workflows — create tasks, assign to agents, track status, mark complete.

Parallel execution coordination — run tasks concurrently, collect results, merge outputs, handle timeouts.

Typed SDK for building FocusMCP bricks — defineBrick() helper with full type safety for tools, manifests, and bus handlers.

Batch file reading — multiple files in one call, deduplication.

Validate data — JSON syntax, JSON Schema compliance, TypeScript type checking, lint rules.

Ops-focused n8n tools for any MCP-compatible client: workflow + execution lifecycle (create, save, archive, delete, cancel, retry single + batch), tag management, security audit, plus composed scanners (find-by-node-type, execution-stats). MCP server + fi

DataNexus MCP — AI-Ready public data intelligence. 35 tools in 7 focused sub-servers: nonprofit, security, compliance, domain, legal, GovCon, and regulatory. Use search_datanexus_tools to find the right tool. No API key required for most tools.

MCP server exposing the full OmniFocus surface to LLM agents — 80 typed tools spanning tasks, projects, tags, folders, perspectives, forecast, review, notes, attachments, and sync, with a strict typed-error taxonomy and per-tool circuit breakers, rate lim

Symbol lookup in source files — find, get, bulk, body.

Cross-references — find who imports or uses a symbol, locate declarations, and trace inheritance chains.

Intelligent file reading — multiple modes to minimize token usage.

MCP server for browser automation using Camoufox - a privacy-focused Firefox fork with advanced anti-detection features

I built a free MCP server for habit tracking — connects to Claude/Cursor It generates complete habit blueprints you can import directly into Focus Streak (Android). 3 prompts that work great: "Help me build a morning routine" → returns 5 importable habits "Why do I keep failing