Best MCP Servers for AI Tool in 2026
1,122 tools indexed · avg score 88.3 · sorted by AgentRank score
The AgentRank index tracks 1,122 ai tool tools. The top-ranked entry is crmne/ruby_llm at 97.9 points. Across the 10 shown here, the average AgentRank score is 88.3, and 10 had a commit in the last 30 days. The most common primary language is TypeScript.
General AI tools cover a broad range of capabilities that enhance language model workflows without fitting neatly into more specific categories. These include AI utility tools, specialized model wrappers, AI-powered content processors, and experimental tools at the frontier of what agents can do.
This is the most diverse category in the AgentRank index — and often where the most innovative tools appear first. New use cases that don't yet have a dedicated category show up here before the ecosystem converges on naming conventions and standard patterns.
When browsing AI tools, the AgentRank score is especially useful as a quality signal since the category is heterogeneous. A high score (especially high dependents + high freshness) is a strong indicator that a tool has found real-world product-market fit, even in a nascent category.
Top 10 AI Tool MCP Servers
- #1crmne/ruby_llm97.9
The Ruby-native AI framework. Chats, agents, tools, images, audio, and video through one consistent API, in plain Ruby or Rails.
Ranks here on adopted by 367 downstream projects and 4.4k GitHub stars, plus a commit within the last week.Dependents 25.3Freshness 23.0Issue health 22.2Stars 11.5Contributors 9.2 - #2AbdelStark/awesome-typesafe-jev94.5
Awesome Jev: a source-backed field guide to TypeSafe's System One model, with SDKs, live demos, agent tools, and independent evaluations.
Ranks here on 513 GitHub stars and a commit within the last week, plus 93% of issues closed.Freshness 30.8Issue health 28.6Stars 13.9Contributors 12.3License 4.6 - #3getkern/kern91.4
a fast, rootless container runtime and sandbox with no daemon. It runs workloads, including agent tool calls and code generated by an LLM, in real containers enforced by the kernel.
Ranks here on 423 GitHub stars and a commit within the last week, plus 100% of issues closed.Freshness 30.8Issue health 30.8Stars 13.5Description 6.2Contributors 5.6 - #4georgeguimaraes/elixir-agent-tools89.3
Elixir development skills with optional Mix checks and Expert language server integration
Ranks here on a commit within the last week and 100% of issues closed.Freshness 30.8Issue health 30.8Stars 11.6Contributors 7.2License 4.6 - #5specula-org/Specula89.0
Specula: An agentic tool for finding deep bugs in system code using TLA+
Ranks here on 490 GitHub stars and a commit within the last week, plus 84% of issues closed.Freshness 30.8Issue health 25.8Stars 13.8Contributors 9.7License 4.6 - #6anvia-hq/anvia86.5
Anvia is a TypeScript runtime for building provider-agnostic agents, tool workflows, and structured extraction inside your application code.
Ranks here on a commit within the last week and 100% of issues closed.Freshness 30.8Issue health 30.8Stars 8.8Contributors 7.2License 4.6 - #7Integral2u/SharpMind85.4
SharpMind. A pure C# / .NET LLM engine — inference, training, and agent tooling in one solution.
Ranks here on a commit within the last week and 100% of issues closed.Freshness 30.8Issue health 30.8Stars 9.3Contributors 5.6License 4.6 - #8joeynyc/awesome-microduck85.0
A curated list of software, simulators, policies, agent tools and coverage for the Pollen Robotics / Hugging Face Microduck robot
Ranks here on a commit within the last week and 100% of issues closed.Freshness 30.8Issue health 30.8Stars 10.8Contributors 5.6Description 4.3 - #9AwesomeHou/dsh-plugin-marketplace83.2
Plugin marketplace for DeepSeek Harness — live-syncs the GitHub dsh-plugin topic (1800+ repos) into a searchable, paginated settings tab with one-click install and agent tools (market_search / mark...
Ranks here on an active commit 14 days ago and 100% of issues closed.Issue health 30.8Freshness 28.3Stars 7.6Description 6.2Contributors 5.6 - #10ccbud/CCBuddy81.0
Coding CLl Buddy, Allow more APIs to access coding cli, protocol conversion, and focus on conversation analysis and debugging agent tools
Ranks here on a commit within the last week and 100% of issues closed.Freshness 30.8Issue health 30.8Contributors 5.6Stars 4.9License 4.6