Chat and assistants
90Topcareer-ops95
Pick a tool you can defend to your team. Every project is scored from its GitHub data and reviewed for what it does well and where it breaks. The rankings update twice a day.
Every project fact-checked on GitHub: stars, last commit, license, Docker files.
Eleven topics, from chat apps to model serving. Each card shows its top project and the categories inside.
Topcareer-ops95
Topdeepseek-harness98
Topmineru94
Topvoicestudio96
TopMoneyPrinterTurbo87
Topn8n95
Topclaw-code97
Toptwenty96
TopOmniRoute96
Topshannon91
Coding agents, skills, starters and the stack you ship on. Paid never moves a rank.
| Rank | Project | Adoption | Freshness | Maintenance | Agent-ready | Score |
|---|---|---|---|---|---|---|
| 1 | deepseek-harnessPlugin-based agent harness from DeepSeek with a local Web UI | 100 | 100 | 100 | 85 | 98 out of 100 |
| 2 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 99 | 100 | 99 | 70 | 97 out of 100 |
| 3 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 96 | 100 | 97 | 85 | 96 out of 100 |
| #4 | codegraphLocal code knowledge graph that coding agents query over MCP | 98 | 100 | 94 | 85 | 96 out of 100 |
| #5 | twentyOpen-source CRM extensible with objects, views and agents defined as code | 97 | 100 | 95 | 85 | 96 out of 100 |
| #6 | gitnexusCode knowledge graph exposed to coding agents through MCP tools | 94 | 100 | 93 | 100 | 96 out of 100 |
| #7 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 97 | 100 | 96 | 70 | 95 out of 100 |
| #8 | skillsAgent skills for planning, TDD, debugging and code review | 100 | 100 | 91 | 70 | 94 out of 100 |
| #9 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 100 | 100 | 90 | 70 | 94 out of 100 |
| #10 | oh-my-claudecodeMulti-agent orchestration plugin and CLI for Claude Code | 91 | 100 | 100 | 70 | 94 out of 100 |
Momentum, Verified build, Docs and Privacy are not measured yet; their weight goes to the signals shown. A dash means the signal is not scored for that kind of project. Hover a number for its rating in words.
Most GitHub stars gained from 17 to 23 Oct 2026, from GitHub's daily star counts.
Claude Code workflow that searches job boards and drafts tailored CVs and cover letters
+15,404 stars this week
Coding-agent skill that runs a multi-phase security audit of a codebase
+12,374 stars this week
Desktop app for voice cloning, dubbing, dictation and audiobooks
+12,201 stars this week
AI apps, agents and models you run, and the parts to build with. Paid never moves a rank.
| Rank | Project | Adoption | Freshness | Maintenance | Agent-ready | Score |
|---|---|---|---|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 97 | 100 | 94 | 85 | 96 out of 100 |
| 2 | OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers | 98 | 100 | 98 | 70 | 96 out of 100 |
| 3 | n8nVisual workflow automation with AI agent nodes and 1500+ integrations | 100 | 100 | 92 | 70 | 95 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 98 | 100 | 89 | 85 | 95 out of 100 |
| #5 | career-opsJob-search agent that scores listings, tailors CVs and drafts applications | 95 | 100 | 94 | 85 | 95 out of 100 |
| #6 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 94 | 100 | 99 | 70 | 95 out of 100 |
| #7 | ai-job-searchClaude Code workflow that searches job boards and drafts tailored CVs and cover letters | 92 | 100 | 98 | 85 | 95 out of 100 |
| #8 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 100 | 100 | 85 | 85 | 94 out of 100 |
| #9 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 99 | 100 | 92 | 70 | 94 out of 100 |
| #10 | mineruDocument parser that converts PDFs, images and Office files to Markdown and JSON | 91 | 100 | 99 | 70 | 94 out of 100 |
Momentum, Verified build, Docs and Privacy are not measured yet; their weight goes to the signals shown. A dash means the signal is not scored for that kind of project. Hover a number for its rating in words.
What each project is made for, from image and voice to trading and science.
Speech to text, text to speech, voice cloning, voice agents and music generation.
Make and edit images, video and 3D, swap faces, upscale, or design graphics and interfaces.
Ask questions of your data, text to SQL, BI and forecasting; read PDFs and scans, extract text and tables.
AI for biology, chemistry, physics and maths, medicine and health, reading papers, robots, and teaching.
Trade stocks and crypto, keep the books, run a shop, market and sell, answer customers, review contracts.
Paying for a closed tool? These are the highest-scoring open-source projects that do the same job, in the category we compare it in. A project marked "own claim" calls itself an alternative in its README.
Nobody pays for a place, and every score shows its parts.
Read the full methodologyStars, last commit, releases, license and Docker files come from GitHub itself, not from the project's own claims.
Each score is built from named signals with published weights. Every project page shows what its score is made of.
Sponsors and tips never move a rank. Spot a wrong fact and you can report it from any page.