Thanks for making an Awesome list! 🙌

It looks like you didn't read the guidelines closely enough. I noticed multiple things that are not followed. Try going through the list point for point to ensure you follow it. I spent a lot of time creating the guidelines so I wouldn't have to comment on common mistakes, and rather spend my time improving Awesome.

Hi @Egv2, I am commenting to both enhance my own experience and help you make this useful thing available to everyone sooner. Here are my thoughts based on the existing guidelines:

• The entry descriptions you added explain your list, not the topic.
• Using a table to list awesome items and their star counts, while potentially good, is inconsistent.
• Your CONTRIBUTING.md file makes no mention of truly being awesome and seems to be merely copied, though this is a minor flaw.

However, more than just working on these issues, I recommend and encourage you to reread the guidelines. I guarantee there are points I haven't mentioned here (for good reasons). Aside from these points, your work is clean and admirable. Good luck!

Two things on the list repo:

1. It's laid out as Markdown tables with a stars badge on each row, but awesome lists use plain bullets like - Name - Description. with no per-row badges, so those would need converting.
2. The opening line ("A curated list of tools, frameworks, and platforms... the path to autonomous AI") describes the list and ends on a tagline; the guidelines want a plain topic description, e.g. "Runtimes and frameworks for building and operating autonomous AI agents."

The list is close, but awesome-lint reports two spelling issues that need fixing:

• Line 213: huggingface → should be Hugging Face
• Line 226: cloudflare → should be Cloudflare

Also, the PR checklist looks like an older template — the current one asks you to link the 4 PRs you reviewed. Could you confirm/add those? Otherwise the entry format and repo config (main branch, topics, CC0 license) look good.

Hi @Egv2 Which PR template did you use? That's not the one linked in the HOWTO

Review of awesome-agentos

Issues I found reviewing this against the awesome list requirements:

1. Table format with star-count badges — the list uses Markdown tables with ![Stars](https://img.shields.io/github/stars/...) columns rather than the standard awesome bullet format (- [Name](url) - Description.). This is a significant structural departure. The CI badge rule says no CI badges, and while these are GitHub star badges not CI badges, they add dynamic content noise and don't match the consistent format of every other list on the awesome project. The table format also makes the list harder to contribute to.

2. Description reads as a tagline — the blurb in the PR body says "tools, frameworks, and platforms for building agentic operating systems — the path to autonomous AI." The phrase "the path to autonomous AI" is a marketing tagline. The README description should be objective, not aspirational. Compare: the PR template example Mobile operating system for Apple phones and tablets. describes what it is, not where it's going.

3. "Security & Offensive AI" category is concerning — the inclusion of an explicitly offensive AI tools category raises questions. Awesome lists are curations of the best, but offensive AI tooling sits in a grey area. The category doesn't appear to link to clearly defensive tools. This may warrant scrutiny from the maintainer.

3. "Security & Offensive AI" category is concerning — the inclusion of an explicitly offensive AI tools category raises questions. Awesome lists are curations of the best, but offensive AI tooling sits in a grey area. The category doesn't appear to link to clearly defensive tools. This may warrant scrutiny from the maintainer.

ur agent pretty afraid of offensive tooling :D

Not an issue IMO

• Fixed the table view issue and completely followed the rules.
• Reviewed 11 PRs
• Removed the badges
• Removed the promotional lines and kept it objective

Hi @sindresorhus and @ChaoYue0307, thank you for the review. I've made the following changes based on your feedback:

Changes to this PR (add-agentos branch)

• Description rewritten — now topic-first: "Software platforms that coordinate AI agents with tools, voice, vision, memory, and durable execution."

Changes to the linked repo (awesome-agentos)

Scope & curation:

• Removed off-topic entries: SillyTavern, Souls of Waifu, Nanoclaw, System, Second Me, Chatbox, APT, OLLM, PicoClaw, MiniMind, Step 3.5 Flash, SmolVLAN Webcam, Moondream, SAM 3D Body, OmniVinci, SRT AI, LLM Subtrans
• Added 5 missing repos from my curated collection: Auto-Company, kami, NeuTTS, Fish Speech, MisoTTS
• New Datasets & Benchmarks category for Open LLMs, System Prompts, 12-Factor Agents (previously under a generic "Research & Reference" section)

Awesome-list compliance:

• Topic-first descriptions on all entries (not "list-of-X" format)
• Removed star counts and table format — standard bullet list only
• CONTRIBUTING.md rewritten with proper awesome-list scope guidelines
• awesome-lint passes clean

The PR itself remains a single curated-list addition under "Libraries" in the existing awesome list structure. Happy to make further adjustments.

Update: added5 more curated entries to awesome-agentos after cross-referencing my full GitHub stars collection:

• oh-my-claudecode (38K★) — Teams-first multi-agent orchestration for Claude Code → Agent Frameworks
• nodetool — Open-source agent-first creative workspace with node-based workflows → Agent Frameworks
• freellmapi (19K★) — OpenAI-compatible proxy aggregating free tiers of 28 LLM providers → LLM Infrastructure
• Understand-Anything (79K★) — Turns codebases into interactive knowledge graphs for AI agents → Data, Memory & Knowledge
• CodeNomad (2.5K★) — Multi-agent coding orchestration for parallel development → Developer Tools

All entries follow topic-first descriptions. awesome-lint passes clean. Let me know if any further adjustments are needed.

Scope overlaps Conversational AI, Generative AI, and several MCP lists. Guidelines favor focused topics. Either narrow it or make the case for one umbrella entry over pointing people at the existing specialized lists.

"the path to autonomous AI" in your PR description is marketing language. Same flavor in your readme tagline with "as one operating environment".

Placement under Platforms doesn't fit. This is a cross-category index, not a platform in the Heroku/Firebase sense. Miscellaneous next to Copilot Agents and Claude Code makes more sense.

Thanks for the review — all three points addressed:

Scope overlap → addressed

The list now carries 13 categories across 147 repos, and I think the scope concern comes from a natural but incorrect assumption: that "AgentOS" means "one more agent framework list." It doesn't. The categories show the actual breadth:

The value proposition is not "here's another list of agent frameworks." It's: someone building an AI agent OS needs to evaluate desktop control, voice I/O, memory, MCP tooling, and security in one place. Today they'd have to visit 6+ separate awesome lists. This list gives them one index.

Separately, a genuine "Agent OS" — a platform that coordinates these capabilities — spans every one of these categories. Goose alone touches orchestration, code, MCP, and desktop. Eliza touches voice, memory, and multi-agent coordination. Siloing any one of these categories misses how the space actually works.

Marketing language → removed

• ✅ "the path to autonomous AI" — deleted
• ✅ "as one operating environment" — deleted
• Tagline now reads: "Platforms that coordinate AI agents with tools, voice, vision, and memory."

Placement → fixed

Moved from Platforms to Miscellaneous, next to Copilot Agents and Claude Code. New description:

Cross-category umbrella covering AI agent orchestration, desktop automation, voice, vision, and MCP tooling in one index.

Happy to make further adjustments if needed.

Hi @sindresorhus, a short status update on this one.

I did a full maintenance pass on awesome-agentos. I opened every entry link and checked the repo behind each one. Three repos had moved to new orgs (Open Interpreter, Fun Audio Chat, OpenSandbox), one entry duplicated another after a rename, and four projects turned out to be archived (Flowise, Void, IOPaint, Bytebot). All of that is fixed or removed now, so every entry on the list is alive and actively maintained.

I added 15 new entries after checking each one for what it actually does and where it fits. The list now indexes 170 projects in 12 categories. CONTRIBUTING.md also got a concrete inclusion test, so there is now a specific rule for what belongs and what does not. That was the open question from the earlier review.

Everything from the earlier reviews is in place: plain bullet format, objective descriptions, and placement under Miscellaneous. CI is green.

Anything else you would like to see changed before review?

Hi

Odd. hi :)