We built Nekko to be the agent you can hand your personal data to and know
it is safe. You choose the models, you choose what data is used, and you
see exactly what is happening, every step.
No account needed to use AI on your computer · Works offline after setup · MIT licensed
MEET NEKKO
SCROLL ▾
WHY NEKKO · FIVE PROMISES
An agent you can trust with your own data
Cloud agents are good. They are also someone else's computer. Nekko gives you the same quality of experience on your own machine, with your data where you can see it.
01 · IN CONTROL
Your models. Your data. Nothing hidden.
Pick the model for every chat: one running on your own computer, or ChatGPT, Claude and Gemini when you want them. The Context Inspector shows every file, note and instruction about to go with your request, so you can leave out what is not needed. Every command the agent wants to run and every line it changes waits for your say.
Local, cloud, or both, chosen per chat and per workflow
See the exact context before it is sent
Approve commands, review changes line by line, undo any of them
The power of a terminal. The polish of a great app.
Tired of the gap between the cloud experience and the local one? We were too. Nekko is a premium app for local AI: the agent, the model server, the diff review, the terminal and your connected tools, together in one window that feels finished. Prefer the shell? The same engine is there as a CLI and an MCP server.
Model server and agent in one app, no second window to babysit
Nekko runs models with llama.cpp under a Rust core tuned for the way agents actually work: repeated edits, many chats open, long contexts. Same model file, same machine, measured the same way as Ollama and LM Studio.
Tokens per second, higher is better Nekko Ollama LM Studio
Plain decode
154
115
116
Code edit
534
106
108
Four requests at once
256
113
210
Gemma 4 12B Q4_K_M, NVIDIA RTX 5090, median of 3 runs, 30 September 2026. Ollama 0.23.0 and LM Studio 2.46.0. Returning to a paused chat took 0.8 s against Ollama's 1.3 s and LM Studio's 1.6 s. Ollama starts a fresh long chat faster (1.3 s to 1.7 s). Your hardware will differ; the harness is open, so run it yourself.
04 · SECURE
A safe space. Even we can't see in.
People need a place where their information stays theirs. We took Apple's idea of a service that cannot see your data even if it wanted to, and built Nekko around it. Local chats never leave your computer. Your phone talks to it through end-to-end encryption, so the relay in between carries only ciphertext. And when you do choose a cloud model, Privacy Shield swaps your emails, keys, card numbers and private names for placeholders before anything is sent, then puts the real values back on the way home.
Local models: nothing leaves your machine, works offline
Phone to computer: end-to-end encrypted, the relay only sees ciphertext
An easy phone app for iPhone and Android. Start a chat, approve a risky command, check on a long run, or dictate a message, all from your pocket while the work runs on your computer at home. No computer nearby? Download a small model and chat on the phone itself, with nothing sent anywhere.
Pair with a QR code, then reach your computer from anywhere
Approvals and questions arrive as a tap, with a buzz
Also in the box: scheduled agents and triggers, Whisper dictation on your computer, cost tracking for cloud models, skills and sub-agents, and connectors for GitHub, Linear, Slack, Discord, Gmail and Drive.
NEKKO BETA · FREE TO JOIN
Help shape what Nekko becomes
Join the beta list and get a front-row seat as Nekko grows. You try new things first, and your feedback decides what we build next.
A chance to try the iOS and Android apps before anyone else
First to test new features as they land
More beta-only extras to come
Psst… want to see what's next?
AI ON YOUR COMPUTER
Your own model server, built in
Nekko comes with its own engine, so running AI on your computer takes a few clicks rather than a weekend of setup. Find a model, see whether it fits your machine, and start using it.
Search and download models from Hugging Face without leaving the app
See which models fit your graphics card before you download them
Load, unload, or keep a model ready so your first reply comes back fast
Serve text and image models on one address that your other apps can use too
Already using Ollama, LM Studio or vLLM? Nekko finds them and manages their models
Add ChatGPT, Claude, OpenRouter or any OpenAI-compatible service alongside
Example setup, not a scan of your machine
Nekko engine● runningbuilt in
Ollama● connectedlocalhost:11434
LM Studio○ not runninglocalhost:1234
PRIVACY AND CONTROL
Share only what you choose
Letting an AI read a prompt is one thing. Handing it your files, your terminal,
your microphone and your network is another. Nekko keeps every one of those
behind a control you set, and shows you exactly what it did.
01
Choose the information it uses
Context means the information sent with your request. Check what is included and leave out what is not needed.
Context InspectorSee the files, instructions, and saved notes included in your next request. Toggle or pin them and see how much space they take up, measured in tokens, the small pieces of text AI models process.
Find relevant filesSearch across the project folders you choose, so Nekko can find useful code and notes without sending every file to the AI.
Guidelines and memoryGive Nekko project instructions through AGENTS.md or CLAUDE.md, and keep saved notes for your preferences. Turn individual instructions and notes on or off.
Save the planTurn the conversation into a project description in SPEC.md, plus a plan and task list that can guide later work.
Improve your requestGet suggestions for making your instructions clearer before you send them. This draft check runs on your computer without an online AI service.
02
Give it useful tools
Connect the tools and repeatable steps your work needs, rather than copying everything into a chat.
Split up a taskDelegate smaller pieces of work to separate AI sessions, called sub-agents. Each gets its own instructions, with limits on further delegation and a view of the work in progress.
Reusable instructionsNot sure which skill to reach for? Nekko's library sorts skills by the job they do, from research and planning to code review and testing, and shows every step before you run one. Install more from the marketplace in a click.
Connect coding toolsMCP is the common language AI tools use to talk to each other. Nekko can use other MCP servers or act as one, so Claude Code, Codex, Cursor or OpenClaw can pass work to your local models through it.
Connect your servicesBring in what matters from GitHub, Linear, Slack, Discord, Gmail and Drive instead of pasting it by hand. Each connection is one you switch on, and it only reaches the network when you use it.
Longer tasks with limitsGive Nekko a goal to plan and work through. Set limits on time or steps, and get a request for guidance if it stops making progress.
03
Review and correct changes
AI can make mistakes. Approval settings, file limits, and change review help you stay involved.
Command approvalsChoose which terminal commands are allowed, need approval, or are blocked. The app checks for risky operations such as deleting files or changing system permissions.
Limit file accessNekko only touches the folders you pick. Your microphone is only used when you start dictating, and speech is turned into text on your computer. For extra peace of mind, Docker can wall off terminal commands entirely.
Review file changesSee edits line by line in the Changes panel. Keep or undo individual lines, whole files, or all changes from the app.
Try again from an earlier stepIf an AI returns an empty reply, Nekko retries and then reports the problem. Use "Reset here" to return to an earlier message and try a different approach.
Meet Nekko: a small pixel cat keeping you company while your agents get to work.
YOUR FILES AND CONNECTIONS
Choose what stays on your computer
ON YOUR COMPUTER
Work without an internet connection
Once your model is downloaded, local work keeps going without the internet. Read, edit and search your folders with nothing sent to an online AI service.
ONLINE SERVICES
Connect online AI when you need it
Online AI models, downloads, and connected services need internet access and may send data to the providers you configure. You choose those connections and are responsible for their costs.
GETTING STARTED
Choose how to use Nekko
Download the desktop app to get started. If you prefer a technical setup, run Nekko in your browser, in Docker, or from a terminal.
The npm package is nekko-agent, publishing with the next release; until then run the CLI from a source checkout (npm run dev or apps/cli). It provides the nekko-agent command.
COMING SOON
What we are building next
Here is what is on the bench right now. Join the beta to try each one first.
Coming soon
Model routers
Let a router pick the right model for each request, so easy questions go somewhere cheap and hard ones get the heavy hitters. Use our upcoming open-source Nekko Router, another open-source router, or bring your own key for a paid one.
Coming soon
Share with friends and teammates
Great work is rarely a solo act. Share agents, workflows and progress with the people you build with, without giving up control of your own machine.
Planned
Control your computer by voice
Dictation is here today. Next, Nekko listens, acts and speaks back, all on your computer: you approve which apps and actions it can use, confirm sensitive steps, and always have a stop button in reach.
Exploring
Run project checks on your machines
A local alternative to hosted CI for repeat jobs such as running tests, with the option of cloud runners for extra capacity. This is not a promise of GitHub Actions compatibility.
READY WHEN YOU ARE
Download Nekko Agent
Free, open-source software. Pick the build for your machine. Model providers and hardware may have their own costs.
Or build from source: git clone · npm install · npm run dev
macOS builds are signed and notarized from v0.6.0 on. An older release is not, and macOS may call it damaged: clear quarantine with xattr -cr "/Applications/Nekko Agent.app", or just take a current build. Read the install guidance
Windows builds are unsigned for now (SmartScreen: More info → Run anyway); free code signing by the SignPath Foundation is in progress.
NEKKO AGENT
Download