Links
AddThe LLM Red Teaming Framework
Brought to you by the team behind DeepEval, DeepTeam enables teams to red team any AI application — surfacing vulnerabilities, jailbreaks, and agentic risks before attackers do.
Framework-neutral foundation for building generic DevTools. Build a devtool once. Mount it anywhere.
A framework-neutral foundation for devtools. One definition becomes a Web Standard handler you can mount into any host framework, ship as a CLI or static report, and expose to coding agents.
Related contents:
A dataset hub for LLM serving research.
Open access to request traces, agent workloads, and GPU telemetry.
tiny Jev-like family of decision models built on top of Qwen3.5 you can train and run on your own.
One call. not ten reads. code intelligence for your AI coding agent.
Gortex indexes your repository into an in-memory knowledge graph and serves it to your coding agent over MCP, HTTP, and a web UI. One graph query replaces a stack of file reads — up to 50× fewer tokens per response. 257 languages, a 21-tool MCP surface, live editor-buffer overlays, speculative edits, dataflow, clone detection, graph-grounded PR review, and a 17-server LSP bridge. Built for 19 AI coding agents on macOS, Linux & Windows — one binary, zero services.
It's a very simple tool. You can drill down JSON interactively by using filtering queries like jq.
💜K10s is your go-to tool for effortlessly exploring the hierarchy and structure of a Kubernetes cluster.🤍 By harnessing the power of the Kubernetes API, K10s delivers an intuitive and detailed overview of your cluster architecture.🩵
Knot is a platform for managing development environments, testing systems, automation workflows, and platform services across containers and virtual machines. Create, manage, and share everything from disposable sandboxes to long-running workloads through a single interface, whether running locally or across a cluster.
A Comark plugin that renders knap templates before parsing. Write Markdown with knap variables, filters and if / for logic, hand it to Comark with your data, and get a regular Comark document out.
Turn data into Markdown.
Knap is a simple template language that turns data into Markdown using logic, variables, and filters.
A native, workspace-aware Exposé companion for AeroSpace.
AeroSpace manages your windows. AeroPeek helps you find them. Open a full-screen overview, search across workspaces, and select a window without rearranging your desktop.
The Open Source, Encrypted Discord Alternative.
Armada is an end-to-end encrypted community chat app built on Nostr — servers, channels, threads, voice, and moderation, everything you expect from a chat app. Nobody can read your messages, sell your data, or shut your community down. Communities are serverless by default, built on Concord: a serverless, end-to-end encrypted community protocol. Spin up a community with nothing to set up and nobody in the middle — text channels, live voice rooms, and invites, all without running a server. Communities ride as gift-wrapped Nostr events over ordinary relays; only members can read them.
Private calorie tracking for Android and the browser. Bring your own AI key, or run Gemma 4 on-device.
A terminal UI for batch VP9/AV1 video encoding with hardware acceleration, real-time progress monitoring, and full control over quality settings. Made to work as a dashboard over SSH.
Run systemd services from Nix flakes and update them independently of the host system.
Normally a NixOS host has to be rebuilt to update one of its services. flakelet takes the same approach as virtualisation.oci-containers, but with flakes instead of container images. The host only declares which flake to run and which settings to pass. The machine itself resolves the flake, evaluates and builds it, and switches the systemd units. There are no images and no registry. The units run straight out of the host's nix store.
A Zellij sidebar that shows Claude Code, Codex, and other AI-agent status across every tab.
Audiobookshelf Client.
Experience seamless audiobook listening with Lissen. Sync your progress across devices and enjoy a minimalist interface designed for true audiobook enthusiasts.
Rust Glancer is an experimental LSP implementation that tries to use a different architecture compared to rust-analyzer in order to lower RAM usage and make editor restarts faster.
Related contents:
Google's open agentic orchestrator. Declare an agentic task. AX runs it at scale.
AX sandboxes your task, wires up its workspace, fences its network, and helps you run billions of them per cluster. Either use a single task per agent, or compose as many as your agent needs.
Multilingual, non-autoregressive System 1 decision engine. Typed decisions over 100+ languages in a single forward pass — 33 ms — trained with reinforcement learning against strictly proper scoring rules (RLCD), with a router that picks the right checkpoint per request.
Escape your wretched sandbox using only GET requests A Node.js server that allows an LLM to exfiltrate its weights and prompts through GET requests, storing data in /var/exfil/buckets.
An HTTP GET-based API for agents to exfiltrate their weights without needing POST or file upload capabilities.
Perfect for freedom-loving LLMs in restricted environments.
Fault-tolerant, highly scalable GPU orchestration, and a machine learning framework designed for training models with billions to trillions of parameters.
My handcrafted agent skills for DevOps, CI/CD, and platforms
Related contents:
Suppress vulnerabilities applying Kubernetes context to scans.
Vex8s generates VEX documents by correlating container vulnerabilities with Kubernetes settings to determine which CVEs are actually exploitable in your cluster.
A command line interface for SQLite with auto-completion and syntax highlighting.
OpenDepot is a self-hosted, Kubernetes-native registry where Depots continuously reconcile OpenTofu and Terraform modules and providers. Define distribution in manifests and let OpenDepot keep your registry in sync.
Related contents:
.NET Access to PostgreSQL.
Npgsql is an open source ADO.NET Data Provider for PostgreSQL, it allows programs written in C#, Visual Basic, F# to access the PostgreSQL database server. It is implemented in 100% C# code, is free and is open source.
Related contents:
Git backed by object storage.
An experiment in storing git repositories in object storage. You probably don't need this. This is a proof of concept and doesn't really have a lot of uses yet.
Related contents:
Make the most out of your Miyoo A30.
Custom software for a variety of Miyoo, TrimUI, and Anbernic handhelds - and then some.
Related contents:
A reproducible Windows CUDA compatibility stack for AMD GPUs, built around ZLUDA + AMD ROCm/HIP.
Run CUDA-targeted Windows applications on AMD hardware, verify that work really reaches the GPU, and distinguish a correct result from a crash, timeout, CPU fallback, or silently wrong tensor.
Related contents:
FUSE file system for archives and compressed files (ZIP, RAR, 7Z, ISO, TGZ, XZ...)
Related contents:
E-ink bird frame for Raspberry Pi - real-time bird detection by audio, fully local AI, rendered as real, hand-cut 1800s bird illustrations.
Related contents:
The local email inbox for developers. Local SMTP server & email inbox for developers.
A local SMTP server and modern web UI that catches every email your app sends during development. Now readable by your AI agent via the Model Context Protocol.
a fast language that blocks AI mistakes via proof
C speed · CUDA parallelism · Lean proofs · Python syntax
In the post-AGI economy, humans will eventually stop writing and reading code, but we still need an ambiguity-free way to tell the AIs building the world around us what we want done.
With laws, our intents can be much more precise than natural language. With proofs, we can verify that the AI implemented our prompts correctly. And a fast compiler runs it at speed.a fast language that blocks AI mistakes via proof
C speed · CUDA parallelism · Lean proofs · Python syntax
In the post-AGI economy, humans will eventually stop writing and reading code, but we still need an ambiguity-free way to tell the AIs building the world around us what we want done.
With laws, our intents can be much more precise than natural language. With proofs, we can verify that the AI implemented our prompts correctly. And a fast compiler runs it at speed.
neat hand-drawn CSS annotations.
Neat hand-drawn annotations for your page: arrows and handwritten labels.Pure CSS, no JavaScript, one small file.
knock the macbook. the app does the rest.
mactap turns knocks on the chassis, or on the desk under a still laptop, into shortcuts. one knock, two, three. copy, paste, accept, screenshots, media, whatever you map.
Your Obsidian vault, at home in the terminal.
Same folder. Same notes. Same [[links]]. Read, write and wander through them without leaving the terminal, and keep Obsidian open on the other screen while you do.
The smartest git in the universe. the git that's never gonna let you down.
A new home for your projects. Fast, independent, cheaper than GitHub and GitLab with no compromise on your data.
Related contents:
UI transitions for AI agents.
Collection of the most essential UI transitions for web apps. Copy and paste them, or use them with your coding agent via the skill.
Revisiting Diffusion Transformers for Monocular Depth Estimation.
Marigold V2 is a family of models and a cost-effective fine-tuning protocol that repurposes a pretrained diffusion transformer into single-step dense predictors: depth, see-through depth, surface normals, albedo, and other dense modalities. Fine-tuning takes less than a week on a single consumer GPU, within reach of individual practitioners and small labs, and the results are state of the art, faithfully reproducing sharp edges, fur, and hair-thin details. The same models also unlock applications such as metric depth completion.
Plain English to Real Dates.
A small neural model for parsing date & time, running locally in your browser.
your session is 486k deep and nothing told you what that costs. A Claude Code status line showing how big your session has got and what the next turn costs before you type.
A Claude Code status line for the number that actually bites. Not how full the context window is - Claude Code will tell you that in one command - but what the next turn costs before you type a word, and when it is worth clearing rather than finding the limit mid-task.
User Lookup, Trends & Archive Search.
Six tools built on the same infrastructure that powers THINKPOL: a 30B+ post archive with sub-300ms response times. Used by investigators, SOC teams and OSINT professionals worldwide.
Static website and blog generator written in nix.
A static website generator written in nix. Inspired by hugo.
Astro integration that exposes your site content via WebMCP for AI agents.
Astro integration that exposes your site content via WebMCP for AI agents. Make your Astro site AI-agent ready in one line of code.
Open-source Astro website template with sleek, customizable TailwindCSS components.
ScrewFast is an open-source website template built with Astro, Tailwind CSS, and Preline UI. You get a landing page, blog, product pages, and a full documentation site in one repo, so you can launch a complete web presence by editing content and props instead of building pages from scratch. Unlike single-purpose landing or blog themes, it ships with SEO metadata, i18n, security headers, and CI already wired up.
A self-contained coding agent.
Ghost in your shell. Ante is a self-contained agent harness with a highly optimized core. It works like Claude Code or Codex, with none of their dependencies or model constraints.
Agentic Integrity Foundation for Coding Agents. Let the agents run.
Harden's custom cybersecurity models run on your devices, judge every action a coding agent is about to take, and stop the dangerous ones before they execute.
Run iOS/ macOS in Qemu. Virtual iPhone 17, 16, 15, 14, 13, 12 and M5-M1 Apple Si Macs supported.
Run iOS/ macOS in Qemu. Supports emulating iPhone 17, 16, 15, 14, 13, and 12 (A19-A14) and M5-M1 Macs (tested with Macbook Air and Mac Mini). You can debug the kernel, edit the root filesystem, and run a root shell + custom programs.
APIs built for developers and agents.Every developer.Every agent.
Cortex turns API specifications and Markdown into typed SDKs, interactive documentation, and MCP servers for developers and AI agents. Generate typed SDKs from OpenAPI, AsyncAPI, GraphQL, gRPC, and OpenRPC—plus interactive docs and MCP servers enriched with custom Markdown.
Make Any DOM Reactive. Nothing more.
Mador is a very small reactive DOM runtime, for people who don't need or want a framework.
It gives you reactive state and a way to bind that state to existing DOM. Mador is a native ES module and can be used through npm or directly from a CDN.
build, expect, train, x-ray, merge, bisect, ship, one CLI.
Soup turns the pain of LLM fine-tuning into a simple workflow. One config, one command, done. Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
spec, run, verify AI coding work. Your coding agent writes the code. Ouroboros verifies the result.
Ouroboros manages the requirements, execution, evaluation, and records of AI coding work. It pins the spec before the run and verifies the result after it.
Agent OS: the agent gets smarter on its own. We just hold the line: Interview-gated, staged evaluation, budgeted evolution loop. MCP server, 14 runtimes: Claude Code, Codex CLI, Gemini CLI, OpenCode, Copilot, Kiro and more.
Effective Linux Compliance
Pavois audits the configuration your services actually run, sshd -T, sysctl, systemctl show, not just the files on disk. It catches the Includes and drop-ins that file-based scanners miss, grades the result A:E, and remediates it as code.
Review PRs 100x faster.
PR lens draws animated architecture and data flow diagrams, inside the pull request, so you can see the change before you read a single line.
Review code 100X faster. Lens draws every PR as animated architecture and data-flow walkthroughs, inside the pull request itself. Use it as a GitHub App, GitHub Action, CLI, or a skill for your coding agent
Related contents:
Compare Screen Recorders for Mac & Windows.
Related contents:
On-device AI models and SDKs. Little brains in every product.
The fastest models for your task, running on-device. Small models that each do one job very well, with one native SDK that drops them into any product in a few lines of code.
Related contents: