Official
Read and write local files with configurable access controls.
filesystem, files, read, write
Best Agent-Friendly MCPs
MCPs designed with autonomous agents in mind: clear tool schemas, well-named actions, predictable failure modes, and tool descriptions that the model can understand without handholding. These are the servers that actually perform when you loop them inside Claude Code, Cursor Agent, or a custom agent harness.
37 MCPs in this collectionFilter:
agent-friendlyTop Agent-Friendly MCPs ranked
Popularity-ordered. Click any card for install snippets, fact sheet, and trust signals.
Official
AI-optimized web search with cleaned, citation-ready results.
search, web, research, rag
Official
Run pre-built browser-automation Actors on managed infrastructure.
browser, automation, scraping, apify
Official
Embedded and hosted vector database for AI agents — open source, zero-ops.
vector-db, chroma, embeddings, retrieval
Official
Persistent memory layer for AI agents — auto-summarised, cross-session recall.
memory, mem0, recall, personalization
Official
Full GitHub API access: repos, PRs, issues, and code search.
git, github, repos, prs
Structured step-by-step reasoning for complex problem solving.
reasoning, thinking, agents, planning
Persistent knowledge graph memory across AI conversations.
memory, persistence, knowledge-graph, agents
Official
Official Microsoft browser automation across Chromium, Firefox, and WebKit.
browser, automation, playwright, testing
Official
Hosted, isolated Chromium runtime for AI agents that need a fresh browser per task.
browser, cloud, browserbase, automation
Official
AI-native browser automation: act, observe, and extract in plain English.
browser, stagehand, browserbase, ai-automation
Official
Managed vector database for semantic search and retrieval in AI agents.
vector-db, pinecone, retrieval, embeddings
Official
Hosted Chromium with proxies and anti-bot evasion for AI agents.
browser, automation, hyperbrowser, proxies
Triage errors, inspect traces, and query events from Sentry.
sentry, errors, monitoring, observability
Official
Open-source vector search with payload filtering, self-hostable or managed.
vector-db, qdrant, retrieval, embeddings
Full GitLab API access: repos, MRs, issues, pipelines, and registries.
git, gitlab, repos, mrs
Capture context and decisions across sessions — durable agent memory in markdown.
agent, memory, context, persistence
Real-time web search with privacy-focused results.
search, web, realtime, news
Full browser automation: navigate, click, screenshot, and scrape.
browser, automation, scraping, puppeteer
Up-to-date library docs pulled directly into your AI context.
coding, documentation, libraries, accuracy
Read and post to Discord channels — community management for AI agents.
discord, chat, community, bot
Search, inspect, and run Hugging Face models and datasets from an agent.
huggingface, ml, models, inference
Query webpages with structured natural language — selectors written for you.
scraping, agentql, extraction, queries
Run any open-source model on Replicate from inside an AI agent.
ai, ml, replicate, inference
Route one prompt across 300+ LLMs with a single API key.
ai, llm, openrouter, inference
Personal task management with due dates — capture, schedule, and complete from chat. Community MCP package archived; Todoist itself is unaffected.
tasks, todoist, scheduling, reminders
AI-driven task management and decomposition for long-running agent projects.
agent, tasks, planning, persistence
Vision-first browser automation built for AI agents.
browser, automation, browser-use, vision
Archived
Secure cloud sandboxes for executing AI-generated code — npm MCP package archived; vendor recommends the hosted E2B platform.
coding, execution, sandbox, cloud
Read and send Slack messages, manage channels and threads.
slack, messaging, communication, automation
Read, create, and manage events across Google Calendar.
google-calendar, scheduling, events, productivity
Vector database with first-class hybrid (vector + keyword) and modular embeddings (MCP server archived — package no longer published).
vector-db, weaviate, hybrid-search, retrieval
Open-source scheduling — booking pages, availability, and event types for AI agents. MCP wrapper is archived; the Cal.com platform itself remains current.
calendar, scheduling, cal-com, booking
Booking-link scheduling — share availability without exposing the raw calendar.
calendar, scheduling, calendly, booking
Read and create events in Apple Calendar via CalDAV from any MCP client.
calendar, apple, caldav, icloud
Archived
Multi-model orchestration — let Claude consult Gemini, GPT, and o-series as sub-agents. Upstream community repository archived; no further maintenance.
agent, orchestration, multi-model, consensus
Open-source agentic browser automation that handles login walls and CAPTCHAs.
browser, agent, skyvern, vision
FAQ: Agent-Friendly
What makes an MCP "agent-friendly"?
Three things: tool schemas that tell the model exactly what an argument is for, action names that match how agents describe intent ("create_issue" not "postIssueV2"), and errors that the model can recover from without a human in the loop.
Do non-agent-friendly MCPs still work with agents?
Usually yes, but they need more babysitting. A poorly-described tool may work for a human-in-the-loop chat but cause hallucinated argument values in an autonomous loop. For agents, prefer the ones on this page.
Are there any MCPs agents should avoid?
Action-taking MCPs without clear confirmation semantics are risky for agents — anything that can send an email, charge a card, or delete data should be carefully scoped. Check the auth model and capability-level filters before wiring into an autonomous loop.