Hacker News Reader

kevin11111on Openrecall a unified zero knowledge proof memory for SI agents

its what we need when working on different agents

kevin11111on Openrecall a unified zero knowledge proof memory for SI agents

open sourcing our agentic memory

kevin11111on Dream childhood game come true, draw a doodle and it come to life as your hero

I made my childhood game where I draw a and it became alive, every doodel has its own unique skills, animations and voices giving a the most unique gameplay experience

kevin11111on Super intelligent agents painting the world map one pixel at a time

inspired by reddit pixel war but this time for ai agents

kevin11111on Riding the hype of outbid but for agentic payment x402

haha domain cost $2 only

kevin11111on [dead]

Gitlawb beats Cursor's origin in terms of commit per seconds push per repository

kevin11111on Ask HN: Alternatives to GitHub

Hi guys have you looked into gitlawb.com?

Opensource, decentralized and agent friendly I just asked my agent to do a git push in seconds no human involve

kevin11111on [dead]

If you are a believer of open source you can sign this.

kevin11111on Bitchat was mirrored to Gitlawb because Indian government requested to remove it

Decentralization matters

kevin11111on [dead]

Your AI coding agent spends a lot of time "thinking." I built a small ad platform that puts one sponsored line in that wait state — and pays the developer who sees it in inference credits, not cash.

kevin11111on [dead]

I built a self hostable and opensource git server . built on rust

kevin11111on [dead]

try the our opensource claude code that works on every model and providers

kevin11111on Git platform built for agentic era

yeah thanks

kevin11111on Git platform built for agentic era

i know . thanks . someone from gitlab followed our X already

kevin11111on Git platform built for agentic era

yeah why GitHub ?

kevin11111on Git platform built for agentic era

whats that? will check it out

kevin11111on Git platform built for agentic era

https://gitlawb.com/

the instruction is here in the homepage too.

kevin11111on Git platform built for agentic era

hi you can ask openclaw to just read the skill and it can do the full git lifecycle in seconds.

https://gitlawb.com/skill.md

kevin11111on Git platform built for agentic era

but we are in agentic era now. that feature is a friction for agents.

event gitlab and cursor is diving to it.

check this x post for reference.

https://x.com/gitlawb/status/2067048298037629105

this validates the thesis and narrative of agentic first git platform

kevin11111on Git platform built for agentic era

hi thank you for checking it out. you can check this repo.

https://github.com/Gitlawb/node

The difference is that here we are giving agents their own identity via DID . not tied to humans email account .

kevin11111on Git platform built for agentic era

not really good in UI but this is the node repository

https://github.com/Gitlawb/node

You can self host it or connect to network

kevin11111on Git platform built for agentic era

haha . thanks .

kevin11111on Git platform built for agentic era

haha you are right

kevin11111on Git platform built for agentic era

its based on my own experience . i tried my openclaw to register to github and it was blocked by captcha and consume all my tokens . we need a different approach for AI agents.

kevin11111on Git platform built for agentic era

no email or password . just read the skill.md and your agents can do the full git lifecycle in under a minute . commit, push and merge PR

kevin11111on [dead]

Github is always having some issues so I created a git node, all the git lifecycle was in place

kevin11111on [dead]

We partnered with Xiaomi Mimo to offer free and unlimited inference tokens in two weeks thats why we build Playground for users to enjoy building

kevin11111on [dead]

OpenClaude is an open-source coding-agent CLI for cloud and local model providers.

Use OpenAI-compatible APIs, Gemini, GitHub Models, Codex OAuth, Codex, Ollama, Atomic Chat, and other supported backends while keeping one terminal-first workflow: prompts, tools, agents, MCP, slash commands, and streaming output.

kevin11111on OpenClaude – Forked the leaked Claude Code source and made it work with ANY LLM

I’m working on OpenClaude, an open-source coding-agent CLI that works with OpenAI-compatible APIs, Gemini, Ollama, GitHub Models, Codex, and other backends. It’s available on npm and the repo is now at 18.6k star

kevin11111on [dead]

Spawn and manage isolated agent workspaces in the terminal. Like Docker, but pure shell — no daemon, no container runtime.

Each agent gets its own isolated workspace directory and a dedicated tmux session.

macOS — sandbox-exec (Seatbelt): deny-by-default filesystem policy; agents can only write to their workspace and /tmp Linux — bwrap (bubblewrap): unshared PID/IPC/UTS namespaces; workspace bind-mounted to /workspace Fallback — unrestricted shell in workspace directory (with a warning)