60 Selected Skills, Workflows, and Open Source Projects, the Ultimate Claude Advancement Checklist

Translation: Peggy, BlockBeats

Editor's Note: If you've been following AI lately, you may easily get the impression that the number of tools is increasing, but the ones that are truly usable are decreasing.

For most people, the challenge lies not in the lack of tools, but in which tools are worth using and where to start.

This list is based on practical testing, systematically reviewing the current AI tool ecosystem, selecting 60 tools, and categorizing them according to different levels and use cases: from coding and development tools, to agent frameworks and tool integration, then to workflow automation, data processing and infrastructure, and finally extending to learning resources and continuous update paths.

Overall, it outlines a clear path from "using AI" to "building AI systems," providing different entry paths for developers, content creators, product builders, and beginners, helping readers find their own entry points in a complex tool ecosystem.

In the ever-evolving landscape of tools, such a structured review may be more meaningful than isolated recommendations. It provides not answers, but a more efficient way to understand and use AI.

Below is the original text:

I've spent over 100 hours testing various AI tools so you don't have to stumble on your own.

Bookmark this first :)

The AI tool ecosystem in 2026 has reached a point of "information overload": new frameworks every week, new agents every day, new projects hitting the GitHub trending list every morning.

Most of them are just hype. Some are indeed useful. A very few will truly change the way you work.

I've filtered out the noise for you.

These 60 tools below are ones I personally tested, selected as truly worth paying attention to—organized by category, along with the scenarios each excels in.

It's recommended to bookmark this, as you are likely to come back to use them repeatedly.

Part 1: AI Programming Agent & IDE

This category of tools can allow AI to help you write code, review code, and even manage the entire development process.
The key is: these are truly capable of running in a real-world workflow, not just staying at the demo stage.

1. Claude Code
A command-line programming agent introduced by Anthropic. It can read files, write code, run tests, all directly in your local environment.
If you want to maintain full control in AI-assisted development, this is the current "gold standard."
https://docs.anthropic.com/en/docs/claude-code

2. Cursor
An AI-first code editor built on VS Code. It supports in-line completion, interaction with code repositories, and multi-file editing.
Suitable for developers looking to seamlessly integrate AI into their existing development flow.
https://www.cursor.com

3. Codex CLI
A terminal-based programming agent introduced by OpenAI. Given natural language instructions, it reads code repositories, generates, and executes code.
Performs well on tasks involving multiple steps.
https://github.com/openai/codex

4. Windsurf
An AI programming IDE launched by Codeium. Its Cascade agent supports multi-file editing, deep understanding of codebases, and an "immersive coding flow."
Growing rapidly.
https://codeium.com/windsurf

5. Superpowers
A collection of 20+ practical skills built for Claude Code, including TDD, debugging, end-to-end pipeline planning, and more.
Over 96,000 stars on GitHub. If you use Claude Code, it is recommended to install this promptly.
https://github.com/obra/superpowers

6.Spec Kit (GitHub)
「Specification-Driven Development」 tool: Write the specification first, then let AI generate the code. Forces you to think before you start coding. Over 50k stars on GitHub.
https://github.com/github/spec-kit

7.Aider
AI pair programming tool in the terminal, supports any LLM. Particularly strong in working with existing codebases. Over 30k stars on GitHub.
https://github.com/paul-gauthier/aider

Part 2: Agent Framework

Used to build automation systems that can "Think-Act-Iterate".

8.OpenClaw
Phenomenal open-source AI agent. Supports long-running processes, multi-channels (WhatsApp/Telegram/Discord), and can even write its own skills. Over 210k stars on GitHub, one of the most beginner-friendly personal AI agent frameworks available.
https://github.com/openclaw/openclaw

9.LangGraph
Orchestrates multiple agents using a "graph structure": supports branching logic, human-in-the-loop, and persistent state.
https://github.com/langchain-ai/langgraph

10.CrewAI
Multi-agent collaboration framework, where each agent has a role, objectives, and a "persona". Ideal for simulating team collaboration processes.
https://github.com/crewAIInc/crewAI

11.AutoGPT
A mature auto agent framework designed for long-running tasks.
Has come a long way since its early versions.
https://github.com/Significant-Gravitas/AutoGPT

12.Dify
An open-source LLM application development platform that integrates workflow, RAG, agent, and model management.
Relatively user-friendly even for non-developers.
https://github.com/langgenius/dify

13.OWL
A multi-agent collaboration framework that excels in GAIA benchmark tests.
Represents the trend of turning cutting-edge research into practical use.
https://github.com/camel-ai/owl

14.CopilotKit
Allows embedding an AI copilot directly into a React application.
Goes beyond just improving development efficiency to making AI an integral part of the product.
https://github.com/CopilotKit/CopilotKit

15.pydantic-ai
A type-safe agent framework based on Pydantic.
Ideal for Python developers looking to produce structured, verifiable results.
https://github.com/pydantic/pydantic-ai

Part 3: MCP Service and Tool Integration

MCP (Model Context Protocol) enables AI to truly "plug into the world." If Skills teach it how to do things, MCP gives it "permission to act."

16. Tavily
A search engine designed specifically for AI agents, providing structured data that can be directly consumed by LLM instead of blue links.
Offers search, extraction, crawling, and mapping tools, with easy MCP integration in one minute.
https://github.com/tavily-ai/tavily-mcp

17. Context7
Injects the latest library documentation into the context of LLM.
No more "API illusion" or deprecated methods.
Simply add "use context7" in your prompt to automatically fetch the latest documentation.
https://github.com/upstash/context7

18. Task Master AI
Your AI project manager. Input a PRD, and it will break it down into tasks with dependencies.
Then, Claude will execute them step by step, transforming a chaotic development process into an orderly pipeline.
https://github.com/eyaltoledano/claude-task-master

19. MCP Playwright
Provides browser automation capabilities to LLM, enabling natural language-controlled real browser interactions for testing, crawling, and interaction.
https://github.com/executeautomation/mcp-playwright

20.fastmcp
Rapidly build MCP services with minimal Python code.
One of the fastest paths to creating custom tool integrations for models like Claude.
https://github.com/jlowin/fastmcp

21.markdownify-mcp
Convert various formats such as PDFs, images, and audio to Markdown.
Enable any document to enter an AI workflow.
https://github.com/zcaceres/markdownify-mcp

22.MCPHub
Manage multiple MCP services over HTTP.
A dashboard to centrally manage all tool connections.
https://github.com/samanhappy/mcphub

Part 4: Claude Skills (Selected)

Skills can inject "professional workflow capabilities" into Claude. Currently, the community has over 80,000 skills, and the following are truly worth installing.

23.PDF Processing (Official)
Supports PDF reading, table extraction, form filling, and file merging and splitting.
One of the most practical skills for knowledge workers.

https://github.com/anthropics/skills/tree/main/skills/pdf

24.Frontend Design (Official)
Used to build a truly usable design system, including bold typography and deployable UI.
Can break away from the common "AI-generated feel" design style. Already has 277k+ installs.

https://github.com/anthropics/skills/tree/main/skills/frontend-design

25.Skill Creator (Official)
A "meta-skill". Describe your workflow in natural language, generate a complete SKILL.md within 5 minutes.
Can create new skills without writing configuration.

https://github.com/anthropics/skills/tree/main/skills/skill-creator

26.Marketing Skills (Corey Haines)
Covers CRO, copywriting, SEO, email automation, growth strategies, and more than 20 skills.
Equivalent to "modularizing" a full marketing team's capabilities.

https://github.com/coreyhaines31/marketingskills

27.Claude SEO
Provides features such as full-site audit, structured data validation, keyword analysis, and more.
Consists of 12 sub-skills, covering the entire SEO workflow.

https://github.com/AgriciDaniel/claude-seo

28.Obsidian Skills
Developed by the Obsidian CEO. Supports automatic tagging, auto-linking, and deep integration with the knowledge base (vault).
If you use Obsidian, this is basically a must-have.

https://github.com/kepano/obsidian-skills

29.Context Optimization
Used to reduce token costs and improve KV cache efficiency.
Can significantly lower API usage costs. GitHub 13,900+ stars.

https://github.com/muratcankoylan/agent-skills-for-context-engineering

30.Deep Research Skill
8-stage research process + auto-writing mechanism.
Suitable for scenarios that require Claude to do "deep research" rather than shallow summarization.

https://github.com/199-biotechnologies/claude-deep-research-skill

Part 5: Local AI and Model Execution

Running models on local devices: higher privacy, faster response, no API costs.

31.Ollama
Run open-source large models locally with a single command. Supports Llama, Mistral, Gemma, and more.
The fastest path from zero to local AI.

https://github.com/ollama/ollama

32.Open WebUI
A self-hosted ChatGPT-like interface. Clean, fast, feature-rich.
Used in conjunction with Ollama, it can build a fully private AI system.

https://github.com/open-webui/open-webui

33.LlamaFile
Packs the entire large model into an executable file.
No dependencies, runnable on download, extremely simple.

https://github.com/Mozilla-Ocho/llamafile

34.Unsloth
Doubles the model fine-tuning speed while reducing memory usage by 70%.
A great starting point if you need to train your own model.

https://github.com/unslothai/unsloth

35.vLLM
High-throughput inference engine, performing 2-4 times faster than traditional solutions.
Currently one of the mainstream choices for open-source model production deployment.

https://github.com/vllm-project/vllm

Part 6: Workflow and Automation

Integrate AI into your existing tools and processes.

36.n8n
An open-source workflow automation tool that supports 400+ integrations and AI nodes, and can be self-hosted.
One of the most powerful visual AI automation builders available.

https://github.com/n8n-io/n8n

37.Langflow
Build agent workflows through a drag-and-drop interface. Over 140k stars on GitHub.
Build complex agent pipelines without code.

https://github.com/langflow-ai/langflow

38.Huginn
Self-hosted Web agents for monitoring, alerts, and data collection.
Emphasizes privacy-first, with all automation running on your own server.

https://github.com/huginn/huginn

39.DSPy
Drive models with "programs" instead of prompts.
A framework originating from Stanford research, suitable for scenarios requiring higher stability.

https://github.com/stanfordnlp/dspy

40.Temporal
A "durable workflow engine" for long-running tasks.
When your automation workflows need to handle crashes, retries, timeouts, and more, this is the standard solution.

https://github.com/temporalio/temporal

Part 7: Search, Data, and RAG

Enabling information to flow in and out of AI systems.

41. GPT Researcher
An automated research agent capable of generating structured reports.
Input a topic, output a full analysis with source references.

https://github.com/assafelovic/gpt-researcher

42. Firecrawl
Convert any website into a format suitable for LLM.
A web crawler tool designed specifically for AI pipelines.

https://github.com/mendableai/firecrawl

43. Vanna AI
Natural language to SQL.
Ask questions in English, directly generate database query statements.
Ideal for those who prefer not to write SQL but need to work with data.

https://github.com/vanna-ai/vanna

44. Instructor
Through Pydantic models, transform any LLM output into structured JSON.
Supports 15+ providers such as OpenAI, Anthropic, Google, etc.
A practical choice for many production-level AI engineers.

https://python.useinstructor.com

45. Chroma
Open-source vector database.
One of the simplest ways to add semantic search and long-term memory to AI applications.

https://github.com/chroma-core/chroma

46.dlt
LLM-focused data pipeline tool that can onboard 5000+ data sources.
Helps you ingest various data into your AI workflows.

https://github.com/dlt-hub/dlt

47.ExtractThinker
Document intelligence ORM.
Extracts structured data from any type of document.

https://github.com/enoch3712/ExtractThinker

Part 8: API and Infrastructure

The "under-the-hood pipeline" that makes everything truly run in production.

48.FastAPI
Python web framework for deploying AI applications.
Extremely well-documented, with built-in Pydantic data validation.

https://github.com/tiangolo/fastapi

49.Portkey Gateway
Accesses 250+ LLM through a single API.
Switch models without changing code.

https://github.com/Portkey-AI/gateway

50.OmniRoute
API proxy supporting 44+ AI providers.
Offers load balancing, failover, and cost optimization.

https://github.com/diegosouzapw/OmniRoute

51.lmnr
Used for tracking and assessing agent behavior.
Allows a clear view of what the agent is doing and measures whether its performance meets standards.

https://github.com/lmnr-ai/lmnr

52.Codebase Memory MCP
Transform your codebase into a "Persistent Knowledge Graph."
Let Claude remember the entire project structure across sessions.

https://github.com/DeusData/codebase-memory-mcp

Part 9: Curated Collection and Learning Resources

Where to continuously acquire information and iteratively evolve cognition.

53.Awesome Claude Skills
Curated skill collection with over 22,000 GitHub stars.
The preferred entry point for discovering new skills.

https://github.com/travisvn/awesome-claude-skills

54.Anthropic Skills Repo
Anthropic's official skills repository.
Also the current standard paradigm for skill building.

https://github.com/anthropics/skills

55.Awesome Agents
A curated list summarizing 100+ open-source agent tools.

https://github.com/kyrolabs/awesome-agents

56.PromptingGuide
An in-depth Prompt engineering guide covering from basics to advanced.

https://www.promptingguide.ai

57.Anthropic Prompt Engineering Tutorial
Contains 9 chapters + Jupyter Notebook hands-on exercises.
One of the best paths for systematically learning about Prompts.

https://github.com/anthropics/prompt-eng-interactive-tutorial

58.SkillsMP
A marketplace platform with a community of 80,000+ skills.
The largest Claude skill directory to date.

https://skillsmp.com

59.MAGI//ARCHIVE
Daily updates on the latest AI project repositories.
Used for tracking cutting-edge advancements.

https://tom-doerr.github.io/repo_posts/

60. Anthropic Official Documentation
Covering all core content including API, Prompt, tool invocation, agent, and more.
If you are serious about building an AI product, it is recommended to read this guide from start to finish.

https://docs.anthropic.com

How to Truly Use This List

Do not try to install all 60 tools at once. It will only lead to information overload and waste of time.

Here is a better approach:

If you are a developer: Start with Claude Code (01) + Superpowers (05) + Context7 (17) + Tavily (16).
This combination can help you set up a powerful AI programming environment with search capabilities and document support.

If you are a creator / knowledge worker: Start with OpenClaw (08) + Obsidian Skills (28) + PDF Processing (23) + Frontend Design (24). This combination can provide you with an AI assistant capable of file management, document processing, and content creation.

If you are working on a product: Start with FastAPI (48) + Instructor (44) + Chroma (45) + LangGraph (09).
This combination covers backend framework, structured output, memory system, and agent orchestration capabilities, enough to support a production-grade AI application.

If you just want to learn: Start with the Anthropic Tutorial (57) + PromptingGuide (56) + Anthropic Official Documentation (60).
Build a solid foundation first, then start adding tools.

Choose a path and dive deep into it first.

Expand your tools slowly as needs arise.

TL;DR

Skills = Teaching AI to "do better"
MCP = Giving AI the "ability to access external tools and data"
Repos = Providing all the open-source infrastructure

Combining these three, what you get is no longer just a "cool-looking demo," but a truly productive AI workflow.

That's it. 60 tools. Now, go build something.

Join the official Coincamps community:

X: https://x.com/coincamps

Telegram: https://t.me/coin_camps