The AI worth using
A curated radar of AI agents, models, tools, MCP servers and skills — kept current by a daily signal. No paywall, ever.
New on the radar
- Hey NoahA proactive AI executive assistant for founders
- AgentSkyAny harness, any LLM — cloud-hosted agents on demand.
- Ctruh StudioCreate interactive 3D and XR experiences without coding
- Airtop for Google Ads AutomationBuild campaigns, optimize spend, and create reports.
- Qwen3.8-MaxQwen’s most capable model for coding and cowork
- DrivenThe trusted AI investment agent, from insight to action
- AtlasoOne memory for every AI you use
- Vibe BuddyHardware for AI coding
- claudemonWild Pokémon appear while you wait for Claude Code
- SnapdownTurn anything on your Mac screen into clean Markdown
- Hand WaveTurn sign language into speech with smart glasses
- VIDEO AI MEMake videos and post them everywhere with just one tool
- mpaiMake existing Codex and Claude Code sessions multiplayer
- MascotAIAnimated SVG mascot studios for apps that need a personality
- PassiveShortsAI Faceless Video Generator for TikTok & YouTube
- ZinleyYour Personal AI Representative for calls, email, and tasks
- CapptivoFree open-source screen recorder & demo editor
- LumichatsA Claude Code alternative for people who avoid the terminal
- UniwebPay SkillFinancial Infra for the AI era
- TermexoA local Windows workbench for Claude Code and Codex
- Bolcho AIBuild Voice AI agents that actually speak India
- trothAn AI partner whose memory lives on your machine, not in a vendor account. Swap the model, keep the mind.
- stepfun-vision-skillCodex Skill：让纯文本模型（DeepSeek）借助 StepFun step-3.7-flash 获得看图能力 | Give text-only Codex models (DeepSeek) image understanding via StepFun step-3.7-flash
- hermes-desktop-achievementsHermes desktop achievements: full achievements page + sidebar nav + statusbar score chip, backed by the bundled hermes-achievements plugin API. Zero new backend
- goal-to-gameBuild beautiful games with high-quality 3D assets using Thrixel and Claude Code.
- pristine-skillAn agent skill enforcing the First-Time Principle: write every change in its final form — no patches, no what-comments, no leftovers, no local/deployed drift. C
- DeepSeek-V4-Flash-0731Frontier agent intelligence at Flash prices
- NudgeForMeAI follow-up agent for missed email opportunities
- Port22Claude Code, Codex & more on your phone
- AgentMicroLive Codex task status in your macOS menu bar
Essential AI tools
- ChatGPTReason, write, and prototype with OpenAI's frontier models.
- ClaudeLong-context reasoning and coding with Anthropic's models.
- GeminiMultimodal reasoning across text, image, and video.
- PerplexityAnswer questions with live, cited web search.
- GrokxAI's assistant, wired into real-time posts on X.
- MistralEuropean frontier and open-weight models, via chat or API.
- DeepSeekStrong open reasoning models at low cost.
- QwenAlibaba's open-weight model family, strong at code and multilingual work.
- KimiMoonshot AI's assistant, built for very long context.
- LlamaMeta's open-weight models you can run and fine-tune yourself.
- CursorAI-native editor that edits across your whole repo.
- GitHub CopilotInline code completion and chat, inside your editor.
- Gemini CLIGoogle's open-source agentic coding assistant, in your terminal.
- Claude CodeAgentic coding in your terminal — plans and edits across files.
- LovableBuild and ship full web apps from a prompt.
- WindsurfAgentic IDE that builds features end-to-end.
- Bolt.newPrompt full-stack web apps that run in the browser.
- ReplitBuild, run, and deploy apps in the browser with an AI agent.
- AiderPair-program with AI from your terminal, committing as it goes.
- ClineAn autonomous coding agent inside VS Code.
- ZedA fast, collaborative editor with AI built in.
- DevinAn autonomous software engineer that takes whole tasks.
- CodexOpenAI's coding agent for the terminal and your IDE.
- AntigravityGoogle's agent-first IDE built around Gemini.
- AmpSourcegraph's agentic coding tool for large codebases.
- OpenCodeAn open-source coding agent that runs in your terminal.
- ContinueOpen-source IDE autopilot you point at your own models.
- Roo CodeAn open-source autonomous coding agent inside VS Code.
- v0Generate production React UIs from a prompt.
- FigmaDesign, prototype, and hand off interfaces in one canvas.
- FramerDesign and publish responsive sites without code.
- shadcn/uiAccessible React components you copy in and own.
- 21st.devBrowse and generate React UI components from a prompt.
- Magic PatternsGenerate and iterate on UI designs from a prompt.
- Builder.ioTurn designs into code and edit it visually.
- WebflowDesign, build, and host production sites without code.
- Tailwind CSSUtility-first CSS for building UI fast in your markup.
- Magic UIAnimated React and Tailwind components for landing pages.
- Cult UICopy-paste animated React and shadcn components and blocks.
- Motion PrimitivesAnimation-first React components built on Framer Motion.
- Watermelon UIReact components built on Tailwind, Radix, and Framer Motion.
- ComponentryAnimated React components built with Tailwind and Framer Motion.
- UiverseThousands of free CSS and Tailwind UI elements to copy.
- Phosphor IconsOpen-source icon family in six weights, including duotone.
- FontshareFree professional-grade fonts from the Indian Type Foundry.
- AnimistaPreview, tune, and copy ready-made CSS keyframe animations.
- AutoAnimateAdd smooth enter, leave, and move transitions in one line.
- Shader GradientConfigure animated WebGL gradient backgrounds and export them.
- Mesh GradientCreate warped shader mesh gradients to export as backgrounds.
- HaikeiGenerate SVG wave, blob, and gradient backgrounds to export.
- Pattern MonsterCustomize repeatable SVG patterns, then copy the CSS or SVG.
- OpenRouterOne API for hundreds of models, with fallback routing.
- GroqRun open models at extreme speed on custom hardware.
- Together AIHost and fine-tune open models via a fast API.
- ReplicateRun and deploy any model with one API call.
- Hugging FaceThe hub for open models, datasets, and demos.
- Fireworks AIFast, production inference for open models.
- FalFast inference for image, video, and audio models.
- ModalRun Python and GPU workloads in the cloud, serverless.
- CerebrasRun open models at record speed on wafer-scale chips.
- BasetenDeploy and serve open models in production.
- DeepInfraRun open models through a low-cost inference API.
- PineconeManaged vector database for fast semantic search.
- SupabasePostgres + pgvector + auth as your AI app backend.
- WeaviateOpen-source vector database with hybrid search.
- QdrantHigh-performance open-source vector search engine.
- ChromaAn open-source embedding database for AI apps.
- LlamaIndexA framework for connecting your data to LLMs.
- NeonServerless Postgres with branching and pgvector.
- TurbopufferServerless vector search built on object storage.
- LanceDBAn open-source vector database that runs on your own storage.
- n8nWire AI into automated workflows, self-hostable.
- ZapierConnect AI to thousands of apps without code.
- MakeVisually automate work across apps and AI.
- CrewAIBuild teams of role-playing agents that work together.
- DifyOpen-source platform to build and run LLM apps and agents.
- LindyBuild AI assistants that automate your busywork.
- OpenAI Agents SDKBuild multi-agent systems with OpenAI's lightweight SDK.
- Agent Development KitGoogle's open framework for building and shipping agents.
- Pydantic AIA type-safe Python framework for building agents.
- smolagentsHugging Face's minimal library for code-writing agents.
- AutoGenMicrosoft's framework for multi-agent conversations.
- LangGraphBuild stateful, multi-step agent workflows.
- InngestRun durable, event-driven background and AI workflows.
- Trigger.devWrite long-running background jobs and AI tasks in code.
- TemporalOrchestrate reliable, long-running workflows as code.
- MastraA TypeScript framework for agents, workflows, and RAG.
- SnykFind and fix vulnerabilities in code and dependencies.
- SemgrepScan code for security bugs with fast static analysis.
- SocketCatch malicious and risky open-source dependencies.
- DopplerManage secrets and environment config across your stack.
- GitGuardianDetect leaked secrets across your repos and CI.
- InfisicalOpen-source secrets management for your team and apps.
- LakeraGuard LLM apps against prompt injection and misuse.
- LangSmithTrace, test, and debug LLM apps in production.
- HeliconeMonitor LLM cost, latency, and usage in one dashboard.
- LangfuseOpen-source tracing and evals for LLM apps.
- BraintrustEvaluate and improve AI products with systematic evals.
- Arize PhoenixOpen-source tracing and evaluation for LLM apps.
- Weights & BiasesTrack experiments, models, and LLM apps in one place.
- SoraOpenAI's text-to-video model for generating clips.
- RunwayGenerate and edit video with AI.
- KlingGenerate cinematic video from text and images.
- LumaGenerate cinematic video and 3D from text or images.
- VeoGoogle's flagship text-to-video model.
- PikaGenerate and restyle short video from a prompt.
- Wispr FlowSpeak naturally and it writes, in your style, anywhere.
- ElevenLabsGenerate realistic AI voices and speech via API.
- SunoCreate full songs, with vocals, from a prompt.
- UdioGenerate music and vocals from a prompt.
- DescriptEdit audio and video by editing the transcript.
- MidjourneyGenerate striking images from text prompts.
- IdeogramGenerate images with reliable, legible text.
- KreaGenerate and enhance images in real time.
- RecraftDesign-grade image and vector generation.
- FluxOpen, high-quality image generation models.
- HeyGenGenerate AI avatar and spokesperson videos.
- SynthesiaCreate AI avatar videos from a script.
- CaptionsEdit and caption talking videos with AI.
- OpusClipTurn long videos into short, captioned clips.
- GammaGenerate decks, docs, and sites from a prompt.
- Wren AITurn natural-language questions into governed SQL and dashboards over your warehouse.
- BrowserOSAn open-source Chromium fork with a built-in AI agent that runs locally.
- OrcaRun several coding agents in parallel, each in its own git worktree, and merge the best.
- StrixAn autonomous agent that finds and validates vulnerabilities like a pentester.
- cogneeGive agents persistent memory backed by a knowledge graph and vector store.
- CactusRun text, speech, and vision models on-device on phones and wearables.
- CartesiaLow-latency text-to-speech built for real-time voice agents.
Models, companies & tools we track
- TypeScriptBuild desktop coding agents using Tauri, SolidJS, and local AI tool integration.
- CodexOrchestrate multi-agent loops to generate editable, publication-ready diagrams via MCP
- Claude CodeUse ancient strategic frameworks to generate structured architectural and product decisions.
- CaliforniaAccess US funding and hardware for India AI projects.
- SwiftFine-tune LLMs locally on macOS using MLX and QLoRA
- GemmaRun Gemma models locally via haxiv on diverse hardware.
- GitHubLock down CI/CD pipelines and block malicious dependencies to prevent supply chain attacks
- NPMSecure CI/CD pipelines and block malicious third-party dependencies.
- GitHub ActionsLock down CI/CD pipelines to block malicious third-party supply chain attacks.
- Simon WillisonTrack forked conversations and de-duplicate databases using content-addressable message IDs.
- GPT-5.6Run complex agent workflows cheaper and faster with frontier reasoning.
- GPT-5.6 LunaTest prompts against local and cloud endpoints instantly from the command line.
- MI300XBypass GPU shortages with alternative high-performance cloud compute.
- GitHub CopilotBuild software faster using integrated AI agents and interactive canvases directly inside your editor.
- MythosStress-test your cryptographic protocols against frontier-model attack vectors before deployment
- iMessageControl local smart-home agents via iMessage on cheap hardware