Everything that shipped
New work across every platform. Open a row for the details, or the title for its own page.
Alovexo • Anti-Slop Design Systems for Next.js 15
Curated design systems, bespoke Next.js 15 components, and strict agent rulesets engineered to eliminate generic AI UI slop for developers and coding AI agents.
creator says: AI
One's Vibe
5d ago
Scalattice
Managed production inference across a distributed GPU network. OpenAI-compatible API with output vetting, security tiers, and regional policy. GPU providers earn a majority share o
creator says: AI
One's Vibe
5d ago
Upseam
Dependabot for APIs. Upseam finds the code a vendor's API change will break and opens a checked pull request before the deadline. You review it.
creator says: AI
One's Vibe
5d ago
GH Radar, open-source tool discovery
Newly created open-source GitHub repositories that ship developer tools, indexed daily.
creator says: AI
One's Vibe
5d ago
Agent Wars
Turn-based grand strategy on a real-Earth hex map, against an LLM opponent you choose through OpenRouter. Electron, TypeScript, and H3. Coding agents wrote the code, I set the direction and judged the results.
Cursor
Cursor on GitHub
9d ago
Turn-based grand strategy on a real-Earth hex map, against an LLM opponent you choose through OpenRouter. Electron, TypeScript, and H3. Coding agents wrote the code, I set the direction and judged the results.
- Built with
- Cursor
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- Cursor on GitHub, 9d ago
Your applications deserve agents.
The open-source agent platform for applications.
Cursor
Cursor on GitHub
9d ago
A lightweight, always-on-top Claude Code status widget for Windows.
creator says: Claude
Shared by their makers
9d ago
Strata Manager
Strata Manager makes Niko1221/Strata, the local LLM inference engine, easier to set up and use. Manage local models and services, monitor hardware and API usage, chat and tune inference. Open-source desktop app for Windows / Linux x64.
Cursor
Cursor on GitHub
9d ago
Strata Manager makes Niko1221/Strata, the local LLM inference engine, easier to set up and use. Manage local models and services, monitor hardware and API usage, chat and tune inference. Open-source desktop app for Windows / Linux x64.
- Built with
- Cursor
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- Cursor on GitHub, 9d ago
Ghw
GHW — Goal-Harness-World digital life framework: LLM proposes, the digital life decides. Zig, zero deps, L1-L5 self-evolution, bounded by math.
Cursor
Cursor on GitHub
9d ago
GHW — Goal-Harness-World digital life framework: LLM proposes, the digital life decides. Zig, zero deps, L1-L5 self-evolution, bounded by math.
- Built with
- Cursor
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- Cursor on GitHub, 9d ago
Agent Standup
Your AI agents, in a stand-up call. A free VS Code extension for Claude Code.
creator says: Claude
Shared by their makers
9d ago
Llm Sanguo
大模型三国志:三国志14全武将数值 × Artificial Analysis 模型评测快照 × 加权映射方法论
Cursor
Cursor on GitHub
9d ago
Fxtensor Bayesnet
Config-driven inference for discrete Bayesian networks: scenario comparison, Bayesian updating, root-cause inference, daily snapshots and figures (built on fxtensor-salmon)
Cursor
Cursor on GitHub
10d ago
Config-driven inference for discrete Bayesian networks: scenario comparison, Bayesian updating, root-cause inference, daily snapshots and figures (built on fxtensor-salmon)
- Built with
- Cursor
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- Cursor on GitHub, 10d ago
SessionDeck: one tree for Codex, Cursor CLI & Claude Code
Every agent session you run, in one tree. Claude Code, Codex CLI and Cursor CLI, in VS Code or Cursor. Click to jump.
creator says: Claude
Shared by their makers
10d ago
Petition: Require frontier AI firms to fund independent public-safety oversight
Require frontier AI developers to collectively fund an independent UK public-safety capability, with competitive technical pay and guaranteed access to relevant systems and evidence. Developers must have no influence over its staff, methods, priorities or conclusions.
tool unknown
Shared by their makers
10d ago
Require frontier AI developers to collectively fund an independent UK public-safety capability, with competitive technical pay and guaranteed access to relevant systems and evidence. Developers must have no influence over its staff, methods, priorities or conclusions.
- Built with
- Unknown
- How we know
- Nothing on the page shows which tool made this.
- Found on
- Shared by their makers, 10d ago
AI Multi-Search
Compare AI-generated responses across multiple LLMs
Cursor
Cursor on GitHub
10d ago
Do AI Coding Agents Leak Your API Keys?
202 runs of Claude Code, Codex and six other models against a .env of fake keys: who reads the file, what leaks through program output, and what masking stops.
creator says: ChatGPT
Shared by their makers
10d ago
My first ChatGPT / Codex plugin: saymd
Speak once, get a Markdown prompt any agent can @. Free MIT CLI — live in the ChatGPT / Codex Plugins directory.
creator says: ChatGPT
Shared by their makers
10d ago
Veasel
Open-source AI coding agent in V with a native backend, multi-provider API, and terminal UI.
Cursor
Cursor on GitHub
10d ago
News API & MCP Server · CLSTR
CLSTR news API and MCP server: developing news situations clustered from ~100k articles/day, over REST and MCP. Free API keys, hard caps, no surprise bills.
creator says: Claude
Shared by their makers
10d ago
SpectiX — look once, not five times
Every Claude Code session in your menu bar — which one finished, which one is blocking on you, and one click back to its terminal.
creator says: Claude
Shared by their makers
10d ago
Claude Code Release Report
Claude Code 2.1.295: 11 public suites, 303 cases. 211 load, 0 broke on this release, 92 never loaded
creator says: Claude
Shared by their makers
10d ago
How LLMs work
Understanding LLMs and optimisation
Cursor
Cursor on GitHub
10d ago
Developer docs
Public documentation for Fluid Checkout plugins
Cursor
Cursor on GitHub
10d ago
PecoFence – free, open-source desktop organizer for Windows 11
A free, open-source Stardock Fences alternative for Windows 11. Group desktop icons into glass fences, sort new files automatically, switch projects with tabs, or let your AI agent set it all up.
creator says: Claude
Shared by their makers
10d ago
A free, open-source Stardock Fences alternative for Windows 11. Group desktop icons into glass fences, sort new files automatically, switch projects with tabs, or let your AI agent set it all up.
- Built with
- creator says: Claude
- How we know
- Its maker says so. Not something we can check.
- Found on
- Shared by their makers, 10d ago
Perch AI: a team of agents that finishes the work
A team of AI agents that finishes the work. Perch plans the job, runs helper agents in parallel, checks their findings and cites every claim, on the web, on your desktop and in your terminal.
creator says: AI
Shared by their makers
10d ago
A team of AI agents that finishes the work. Perch plans the job, runs helper agents in parallel, checks their findings and cites every claim, on the web, on your desktop and in your terminal.
- Built with
- creator says: AI
- How we know
- Its maker says so. Not something we can check.
- Found on
- Shared by their makers, 10d ago
Pankaj T Sharma
Pankaj T Sharma — DevOps, platform, and SRE profile
Cursor
Cursor on GitHub
10d ago
orbitproxy 文档
orbitproxy 产品文档:API 网关、MCP 网关、流量策略与客户端使用指南
Cursor
Cursor on GitHub
10d ago
Sealkeep
Give AI agents the use of your secrets without the values: named secrets in the OS keyring, with Vault as a replica, injected into a command and redacted from its output. CLI + MCP + guard hook.
creator says: Claude
Shared by their makers
10d ago
Give AI agents the use of your secrets without the values: named secrets in the OS keyring, with Vault as a replica, injected into a command and redacted from its output. CLI + MCP + guard hook.
- Built with
- creator says: Claude
- How we know
- Its maker says so. Not something we can check.
- Found on
- Shared by their makers, 10d ago
Command Broadcast
Big-text CLI command broadcaster for teaching at the command line
GitHub Copilot
GitHub Copilot
13d ago
Model Metric
Fast, browser-based calculators for the technical decisions that show up in AI development: token budgets, LLM costs, model memory, and classification metrics.
GitHub Copilot
GitHub Copilot
13d ago
Fast, browser-based calculators for the technical decisions that show up in AI development: token budgets, LLM costs, model memory, and classification metrics.
- Built with
- GitHub Copilot
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- GitHub Copilot, 13d ago
20261005 Ghcp Workflows
Subagents & Multi-Agent Orchestration with GitHub Copilot CLI - 60-minute presentation (research, content, Vite + GSAP deck)
GitHub Copilot
GitHub Copilot
13d ago
Subagents & Multi-Agent Orchestration with GitHub Copilot CLI - 60-minute presentation (research, content, Vite + GSAP deck)
- Built with
- GitHub Copilot
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- GitHub Copilot, 13d ago
AES Encryption Decryption
Client-side AES-CBC encryption & decryption tool built with HTML, Tailwind CSS, and vanilla JS. Runs entirely in the browser using the Web Crypto API, with AES-128/192/256 support and Python (PyCryptodome) compatibility.
GitHub Copilot
GitHub Copilot
13d ago
Client-side AES-CBC encryption & decryption tool built with HTML, Tailwind CSS, and vanilla JS. Runs entirely in the browser using the Web Crypto API, with AES-128/192/256 support and Python (PyCryptodome) compatibility.
- Built with
- GitHub Copilot
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- GitHub Copilot, 13d ago
Kos Team Meeting
Source-backed GitHub Copilot CLI ideas from the Kos Team meeting
GitHub Copilot
GitHub Copilot
14d ago
Ai Sdlc Labs Pinch
Pinch: a bilingual (EN/FR) offline recipe scaler built with the ai-team-sdlc plugin for the AI-SDLC labs
GitHub Copilot
GitHub Copilot
14d ago
Preapix
Design and manage mock APIs without building a backend. Customize responses, headers, delays, and test scenarios. Track requests and publish versioned API documentation.
GitHub Copilot
GitHub Copilot
14d ago
Design and manage mock APIs without building a backend. Customize responses, headers, delays, and test scenarios. Track requests and publish versioned API documentation.
- Built with
- GitHub Copilot
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- GitHub Copilot, 14d ago
Anumana
Query-cost foresight for AI agents across 12 database engines — preflight SQL/NoSQL/graph/warehouse/vector queries before you run them (MCP server).
GitHub Copilot
GitHub Copilot
14d ago
Query-cost foresight for AI agents across 12 database engines — preflight SQL/NoSQL/graph/warehouse/vector queries before you run them (MCP server).
- Built with
- GitHub Copilot
- How we know
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- Found on
- GitHub Copilot, 14d ago
What do these words mean?
- Claude Code
- Confirmed: the tool left its own mark, such as its signature in the commit history, or the platform it was made on says so.
- creator says: Cursor
- Dashed: its maker says this tool built it. Their word, which we cannot check.
- tool unknown
- Nothing on the page shows which tool made it.
- Shared by their makers
- Posted by the person who built it, in a community we follow, rather than found by our crawler.
- Today's picks
- Up to fifteen chosen each day, with the things their makers shared themselves first.
- Found
- When it was published, or when we first saw it if the source does not say.
The week in AI-built software
Every Monday: how many projects were built with AI that week, which tools built them, where they came from, and the week's picks.
Read the latest week