FileCity

112 posts

FileCity banner
FileCity

FileCity

@FileCityAI

The fastest way to understand any codebase. Interactive. Visual. Narrated step by step. Part of https://t.co/ttaQULU4dv

Se unió Ocak 2026
0 Siguiendo176 Seguidores
Tweet fijado
FileCity
FileCity@FileCityAI·
You know what's better than a README? A tour guide. Introducing File City Tours—the fastest way to understand any codebase. Interactive. Visual. Step by step. We love the work @mrlesk does and wanted to showcase his Backlog.md. We forked it and turned it into an explorable city. Check the map first: 845 files/79 folders; Markdown clusters in blue dominating. #DevTools #AIAgents
FileCity tweet media
English
3
3
18
3.8K
Better Auth
Better Auth@better_auth·
Today we're announcing Agent Auth Protocol An open standard for agent authentication, capability based authorization and service discovery ⇃read more ⇂
Better Auth tweet media
English
34
69
879
73.4K
FileCity
FileCity@FileCityAI·
@404Cause @imbue_ai @modal Its on our roadmap, but if you have a repo in mind we could use as guidance that would be useful. We are creating this file city concept to couple with our telemetry manifest and types feel like an important part of a codebases core manifest
English
0
0
1
26
Danver Braganza
Danver Braganza@404Cause·
@FileCityAI @imbue_ai @modal Hey @FileCityAI, I'm one of the developers of Offload here. Nice work, thanks for providing this visualization and walkthrough! Have you thought about visualizing the dependencies between the Rust types as well?
English
1
0
2
45
Imbue
Imbue@imbue_ai·
Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run
English
14
16
176
65.6K
FileCity
FileCity@FileCityAI·
Made you a File City Tour of cookbook A visual map + narrated walkthrough of the collection of examples, end-to-end tutorials, and ready-to-run apps built with Liquid AI Foundational Models (LFM) and the LEAP SDK for fully local, on-device AI on laptops, mobile, edge, and browsers github.com/Liquid4All/coo… Great work! @paulabartabajo_ @liquidai
English
0
0
4
107
FileCity
FileCity@FileCityAI·
Made a File City Tour of Entropic A visual map + narrated walkthrough of the secure local-first 1-click AI workspace that runs OpenClaw agents in isolated containers with armored security, malicious skill scanning, and full data privacy. github.com/dominant-strat… Great work! @Entropic_AI
English
0
0
6
1.1K
Erick
Erick@ErickSky·
🚨 ¡BOMBA EN LA IA LOCAL! @Entropic_AI acaba de hacer OPEN SOURCE todo el código de Entropic: el primer workspace de IA seguro, local y de 1 clic para tu máquina principal. Esto es HISTÓRICO porque: ✅ Corre @openclaw de forma aislada y segura. ✅ Local-first: tus datos y herramientas NUNCA salen de tu PC. ✅ Security-first: contenedores blindados + escáner de skills maliciosos. ✅ Instalación en 1 clic (Mac & Linux ya, Windows pronto). ✅ 100% extensible: crea, modifica y lleva tu agente a otro nivel. Olvídate de la nube, de los riesgos de privacidad y de los setup eternos. Tu máquina. Tus datos. Tu control total. REPOOO👇
Español
24
189
1.2K
60.9K
Rivet
Rivet@rivet_dev·
Introducing the Secure Exec SDK Secure Node.js execution without a sandbox ⚡ 17.9 ms coldstart, 3.4 MB mem, 56x cheaper 📦 Just a library – supports Node.js, Bun, & browsers 🔐 Powered by the same tech as Cloudflare Workers $ 𝚗𝚙𝚖 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 𝚜𝚎𝚌𝚞𝚛𝚎-𝚎𝚡𝚎𝚌
English
44
60
820
232.8K
Luke The Dev
Luke The Dev@iamlukethedev·
@FileCityAI This is brilliant and golden! Can you do one when I upload the entire code?
English
1
0
3
105
Luke The Dev
Luke The Dev@iamlukethedev·
Claw3D repo is live. No code yet. But the foundation is there. If you’re interested in the vision: ⭐ Star the repo 💬 Drop ideas in Discussions Code drops this week. Let’s build the future of AI workspaces together. 👉 github.com/iamlukethedev/…
English
59
26
253
19.7K
FileCity
FileCity@FileCityAI·
Made a File City Tour of Crucix A visual map + narrated walkthrough of the open-source personal intelligence terminal / Jarvis-style local dashboard that pulls real-time data from 26+ OSINT sources (satellite fire detection, flight tracking, radiation monitoring, conflicts, markets, etc.). github.com/calesthio/Cruc… Great work! @calesthio
English
0
0
8
606
God of Prompt
God of Prompt@godofprompt·
🚨 Holy shit… Someone just open-sourced a personal intelligence terminal that pulls from 26 data sources in parallel every 15 minutes. Satellite fire detection. Flight tracking. Radiation monitoring. Conflict data. Live markets. Sanctions lists. Social sentiment. All rendered on a single Jarvis-style dashboard running locally on your machine. → NASA FIRMS satellite imagery → Real-time ADS-B flight tracking → FRED economic indicators → Armed conflict event mapping → Nuclear radiation monitoring → Maritime vessel tracking → 17 Telegram OSINT channels Hook it to Claude, GPT, or Gemini and it becomes a two-way intelligence assistant. Push alerts to Telegram and Discord. Run commands like /brief and /sweep from your phone. Zero cloud. Zero subscriptions. One Express dependency. Just node server.mjs. The repo is called Crucix.
God of Prompt tweet media
English
17
82
615
39.3K
Nav Toor
Nav Toor@heynavtoor·
🚨 NVIDIA just solved the biggest problem with AI agents. They keep breaking things. It's called NemoClaw. One command to install. Your AI agent now runs inside a fully locked-down sandbox. No more agents deleting your files. No more leaked API keys. No more unauthorized network requests. Here's what it locks down: → Network: blocks all outbound connections unless you approve them → Filesystem: agent can only access /sandbox and /tmp → Process: blocks privilege escalation and dangerous system calls → Inference: all model calls routed through controlled backends If the agent tries to reach a website you haven't approved, it gets blocked. You see the request and decide. Here's what's inside: → NVIDIA Nemotron models pre-installed (120B and 30B) → Runs locally, on NVIDIA cloud, or through vLLM → Telegram bridge for chatting with your agent from your phone → Works with any OpenClaw or coding agent → One-click deploy to remote GPU through Brev Announced at GTC 2026. Built by NVIDIA. That's it. Your AI agent now runs in a cage. 100% Open Source. Apache 2.0 License.
Nav Toor tweet media
English
21
21
110
8.8K
FileCity
FileCity@FileCityAI·
Made you a File City Tour of Velxio A visual map + narrated walkthrough of the free and open-source Arduino emulator that runs entirely in the browser with AVR8 & RP2040 emulation, 48+ electronic components, code editor, local compilation, and Docker self-hosting. github.com/davidmonterocr… Great work! @CamilleRoux
English
0
0
5
96
Camille Roux
Camille Roux@CamilleRoux·
Velxio : un émulateur Arduino gratuit et open source qui tourne entièrement dans le navigateur. Émulation AVR8 et RP2040, 48+ composants électroniques, éditeur de code, compilation locale. Peut être auto-hébergé avec Docker velxio.dev
Français
6
32
157
8.7K
FileCity
FileCity@FileCityAI·
@yazins Made you a File City Tour of OpenOats A visual map + narrated walkthrough of the open-source AI meeting copilot for macOS with meeting templates and real-time structured AI notes. github.com/yazinsai/OpenO… Great work! @yazins
English
1
0
11
425
yazin
yazin@yazins·
Just shipped the biggest update to OpenOats (prev OpenGranola: the open-source meeting copilot for macOS): Meeting Templates + AI Notes ✨ Pick a template before your meeting (1:1, Customer Discovery, Hiring, Stand-Up, Weekly) and after it ends, generate structured notes from the full transcript with one click. Notes stream in real-time as the LLM writes them. Browse past sessions, regenerate with different templates, copy to clipboard. Also fixed a subtle bug where the last ~3 seconds of audio could be lost when stopping a session. Now the app gracefully drains all buffered speech before closing. 4 new files, 1,277 lines added. All open source. github.com/yazinsai/OpenO…
yazin tweet media
yazin@yazins

OpenGranola now runs 100% locally with Ollama. LLM suggestions, knowledge base embeddings, transcription — all on your Mac, nothing hits the network. Just point it at Ollama instead of OpenRouter in settings and you're done. github.com/yazinsai/OpenG…

English
51
16
342
53.4K
GitHub Projects Community
GitHub Projects Community@GithubProjects·
NVIDIA NemoClaw is an open source stack that simplifies running OpenClaw always-on assistants safely. It installs the NVIDIA OpenShell runtime, part of NVIDIA Agent Toolkit, a secure environment for running autonomous agents, with inference routed through NVIDIA cloud.
GitHub Projects Community tweet media
English
14
25
209
15.2K
lidang 立党 （全网劝人卖房、劝人学CS、劝人买SP500和NASDAQ100第一人）
lidang 立党 （全网劝人卖房、劝人学CS、劝人买SP500和NASDAQ100第一人）@lidangzzz·
我开源了一款非常简单但是极度好用的multi-agent system。 名字就叫goal-driven（目标驱动） github.com/lidangzzz/goal… 原理非常简单，就像我说过的，让一个master agent不断监督subagent工作，直到subagent完成了工作，并且master agent亲自认证工作meet the criteria，符合判据中的要求， 否则就继续无限循环，逼迫subagent持续工作。 顾名思义，goal-driven就是goal-driven，就是一个纯粹目标驱动的框架，只要根据criteria判据判断达不到目标，就必须收集一切错误信息进行改进，进行无限循环，直到彻底实现最终goal（目标）为止。 这个设计的优势是，可以让你以>100小时的时间、极高的token消耗成本，去让这个master agent-subagent system去持续解决设计复杂、逻辑复杂、高度抽象、挑战人类和AI Agent能力极限的极其困难的问题， 比如设计编译器/interpreter/transpiler、解决复杂的数学问题、复杂系统问题、电子仿真问题等等， 目前这个goal-driven已经完成了三个工作， 一个是全自动设计了sqlite的一个rust实现版本，sql parser非常稳定， 一个是全自动设计了C++实现的TypeScript Compiler，这在去年只有微软几个人实现了这个工作， 一个是全自动实现了世界上最复杂的数学证明工具——lean4的TyeScript版本编译器，这在过去也是由微软发明并且由整个数学学术界维护的。 goal-driven已经被证明了是一款非常强大、非常简单、非常直白的multi-agent system，而且只需要一个prompt写进去即可， 只要你设定好了goal（目标），criteria（判据，判断成功与否的标准，比如生成1000个复杂test case）， 就可以让agent和subagent持续工作100小时，完成一些人类历史上极具挑战难度的、最复杂、 最抽象、最艰难的工作。 好了，赶紧点个star，接下来我还要开源一个更重磅的东西。 github.com/lidangzzz/goal…
中文
33
59
346
80.3K
🍓🍓🍓
🍓🍓🍓@iruletheworldmo·
bookmark this immediately. someone just open sourced the methodology that changes how ai agents write code. 40,000 stars on github in days. it's called superpowers and if you're using claude code or codex without this, your so cooked. don't fall behind.
Ihtesham Ali@ihtesham2005

🚨 Holy shit...A developer on GitHub just built a full development methodology for AI coding agents and it has 40.9K stars on GitHub. It's called Superpowers, and it completely changes how your AI agent writes code. Right now, most people fire up Claude Code or Codex and just… let it go. The agent guesses what you want, writes code before understanding the problem, skips tests, and produces spaghetti you have to babysit. Superpowers fixes all of that. Here's what happens when you install it: → Before writing a single line, the agent stops and brainstorms with you. It asks what you're actually trying to build, refines the spec through questions, and shows it to you in chunks short enough to read. → Once you approve the design, it creates an implementation plan so detailed that "an enthusiastic junior engineer with poor taste and no judgement" could follow it. → Then it launches subagent-driven development. Fresh subagents per task. Two-stage code review after each one (spec compliance, then code quality). The agent can run autonomously for hours without deviating from your plan. → It enforces true test-driven development. Write failing test → watch it fail → write minimal code → watch it pass → commit. It literally deletes code written before tests. → When tasks are done, it verifies everything, presents options (merge, PR, keep, discard), and cleans up. The philosophy is brutal: systematic over ad-hoc. Evidence over claims. Complexity reduction. Verify before declaring success. Works with Claude Code (plugin install), Codex, and OpenCode. This isn't a prompt template. It's an entire operating system for how AI agents should build software. 100% Opensource. MIT License.

English
47
78
709
150.2K
Brace
Brace@BraceSproul·
We're re-releasing Open SWE with a new emphasis on deep integrations with the applications you already use It integrates with: - Slack - Linear - Github so you can use it directly from the applications you already spend your time in, without needing to learn a new platform Checkout our announcement blog, and use the open source agent today: github.com/langchain-ai/o…
LangChain@LangChain

x.com/i/article/2033…

English
5
12
147
39.7K
Arlan
Arlan@arlanr·
we built a framework called crust to make it easier to build CLIs for ai agents we’re also on the front page of hacker news
Arlan tweet media
English
6
3
68
7.4K
Chao Huang
Chao Huang@huang_chao4969·
Introducing ClawTeam: Agent Swarm Intelligence 🚀 ( github.com/HKUDS/ClawTeam ). The Evolution of AI Agents: Solo 🤖 → Swarm 🦞🤖🤖🤖 AI assistants like OpenClaw and nanobot have made it incredibly easy for everyone to have their own personal agents. They're everywhere now — coding, writing, analyzing. But here's the thing: they're all working in isolation. It's like having a bunch of brilliant interns who never talk to each other. We think it's time for the next leap. ClawTeam transforms those isolated agents into collaborative swarms that actually think and work as a team. No more babysitting multiple agents or juggling contexts. Just tell the leader agent your goal — it spawns the right specialists, divides work intelligently, and orchestrates everything until completion. It's like upgrading from solo freelancers to a synchronized dev team that never sleeps. ⚡ From Hours to Minutes, From Complex to Simple Here's where it gets interesting: whether you're running ML experiments across 8 GPUs, building full-stack applications, or analyzing market data, ClawTeam turns complex multi-day projects into single-command operations. We're not just making agents faster — we're unlocking collective intelligence to tackle something big. #ClawTeam #OpenClaw #nanobot #AIAgents
Chao Huang tweet media
English
53
169
890
54.3K
Drey
Drey@dr3ynow·
A couple weekends ago I attended the @sotalikesfuture x @ARIA_research hackathon. Challenge 1: build a trustworthy multi-agent network. Agents that can identify each other, prove who sent what, delegate tasks with limits, and reject anything outside the trusted network. The challenge mapped almost perfectly to what we'd been building at Kanoniv. Identity and trust for AI agents. The hackathon didn't inspire the idea, it confirmed the ecosystem needs it now. So we shipped it. Today we're open-sourcing kanoniv-agent-auth. Every agent gets its own identity and keypair. Agents can delegate tasks to sub-agents with scope and budget caps, and authority narrows at each level, never widens. Every tool call is signed and verified before it executes. Revoke an agent's access and it takes effect immediately, no unwinding. Full audit trail on every action. Who did what, who authorized it, how deep in the chain. Rust, TypeScript, Python. Same behavior across all three. MIT licensed. github.com/kanoniv/agent-… Shoutout to @ObadiaAlex and the team for organizing. The challenge was exactly the right question at the right time. If you're building with agents that call tools, spend money, or talk to other agents, try it, tell me what breaks.
English
5
8
38
3.2K

Descubrir

@better_auth @himseIf_65 @404Cause @imbue_ai @modal @paulabartabajo_ @liquidai @Entropic_AI