Orynr Code Help

Everything you need to get going with Orynr Code: installing it, connecting a model, working with the agent team, keeping projects safe, and fixing anything that misbehaves.

Installing

Orynr Code is free. It runs on Windows 10 and 11 (64-bit) and installs for your user only, so it needs no administrator rights and nothing else: no Python, no Node.js.

1. Download orynr_code_setup.exe from the Orynr Code page. Only download it from orynr.com.
2. Run it. Windows may show “Windows protected your PC” the first time; click More info → Run anyway. The file is signed by Orynr LLC.
3. Read and accept the licence, choose the folder (the default is fine) and tick desktop shortcut if you want one.
4. Finish. Start it from the shortcut, or open a new Command Prompt or PowerShell window and type ora.

Check your download (optional)

Every release lists its SHA-256 fingerprint on the download section. In PowerShell, in the folder with the file:

(Get-FileHash .\orynr_code_setup.exe).Hash

If the result matches the published value, your copy is exactly the one Orynr released.

Your first chat

When Orynr Code starts it shows “Please wait while we get our services ready.” for a few seconds, then “Ready. Press Esc to use Orynr Code.” Next it asks which project folder to work in:

• Continue where you left off, New project (a folder in Documents\Orynr_Projects), a recent project, or Browse to any folder on your PC.
• Later (or Esc) keeps the chat in the projects folder for now. Say what you want to build and the Agent suggests a short name (like online-store), asks if it's OK, and creates the project.
• /new always asks for the folder again, so every new chat starts in the right place.
• From a terminal you can also open a folder directly: ora C:\path\to\project.

Connecting a model

Orynr Code doesn't come with an AI model; you choose one. It can be a cloud provider with your own key, or a model server you run yourself.

1. Open the model picker with ctrl+alt+o (or /models).
2. Press ctrl+alt+a for Connect a provider (or /connect).
3. For your own server, pick Local or network server (top of the list). For a cloud service, pick it from the list and paste your API key.
4. Choose a model from the picker. It shows up at the bottom of the prompt.

See Cloud providers and Your own model server for details.

Keys & commands

Every shortcut is ctrl+alt plus a letter. Not sure? Type /help: it lists every command with its shortcut and you can search it. The bottom line of the screen reminds you: /help command for help.

Inside the input box the usual text keys work as in any editor (ctrl+a/ctrl+e start/end of line, ctrl+w delete a word, arrows, Home/End).

The agent team

The default Agent (Orynr) does the work itself, with every tool: it reads and edits files, runs commands, uses the built-in tools, and checks its own work. For bigger jobs it can start helpers that work alongside it in the background and always report back. Press Tab to pick a mode yourself.

Talking while it works

You don't have to wait. Type a new message while the agent is busy: it pauses, it doesn't stop. It finishes the step it is writing, the actions it had already decided and the command that is running, then reads your message and answers it. Nothing it did is thrown away or done twice. Helpers keep working in the background and report when they're done.

If the model hadn't started answering yet (for example it was still waiting for a busy server), your message is read straight away.

Press Esc to stop everything at once.

Approvals & your folder

Before it runs a command or changes a file, Orynr Code asks. Here's how to answer.

• Allow once: just this action.
• Allow always: this kind of action from now on. The screen lists exactly what that covers before you confirm: for a command it's the program (for example git status *, which does not include git push), for file edits it's any file.
• Reject: it won't do it. You can tell it what to do differently.

“Allow always” never lets a chained command slip through: cd src && rm -rf build still asks unless every part of it was allowed, and a command hiding another command inside it only runs as exactly approved.

The project folder rule

• Inside the project folder the agent reads and writes freely (within your approval setting).
• Reading anything outside it needs your OK once per chat (Allow for this chat).
• Changing anything outside it, or running a command that names an outside path, is blocked.

How often it asks

/permission switches between Ask Before (start here), Auto Approve, Auto Decline and Full access.

Reviewer & to-do list

While you work, a reviewer checks what the agent did against what you asked, its own to-do list and what it said it would do, and hands it whatever was missed. The agent sees its open to-do items and the reviewer's notes with every message, whichever agent you're on, so nothing quietly gets dropped.

• The reviewer works between the agent's steps, never on top of them, so it doesn't slow your chat down on a single-GPU model server.
• Items that need you (a password, a decision) are listed separately as waiting on you; the agent mentions them once instead of retrying.
• The side panel shows the current to-do list and the reviewer's open items.

Undo & history

• Undo: before each message, your project files are snapshotted. /undo puts them back to that point (it works even if the project doesn't use git); /redo goes forward again.
• Sessions: every chat is kept per project. ctrl+alt+l lists them.
• Long chats are condensed automatically near the model's limit, so they never overflow. /compact does it now.
• Share a chat as a read-only page served from your own machine (/share, /unshare), or /export it.

Long jobs & retries

Orynr Code is built so a job never dies or goes quiet halfway.

• Long commands are fine. Installs, builds and test runs may take up to 30 minutes.
• “Timed out” isn't “failed”. If something is still running at its limit, it's stopped along with everything it started, and the agent checks what actually happened (is Python installed now? do the files exist?) before doing anything again.
• A model that doesn't answer is tried again, with a pause between tries, and you see it on screen: “trying again, attempt 2 of 6”.
• After 6 failed tries it tells you plainly what it was doing and what it found, instead of looping.
• A busy model server (for example a GPU box finishing someone else's job) is waited on for up to 30 minutes, with a “the model server is busy” line. Esc stops the wait any time.
• While your own server reads a long chat you see how far it has got: “Reading the conversation — 32,000 of about 41,000 tokens”.

Switching model on a long chat

A model can't use another model's memory of a chat, so the model you switch to has to read it. To keep that quick, the new model gets all your messages, all its replies, every action and the latest work in full, while older tool results (file dumps, logs) become one-line notes. Nothing is deleted: the model brings any of them back with recall when it needs it. On a real 300,000-token chat, the new model read about 41,000.

Cloud providers

Orynr Code knows 75+ providers and thousands of models, including OpenAI, Anthropic, Google, DeepSeek, xAI, OpenRouter, Groq, Mistral, GitHub Copilot, Azure and Amazon Bedrock.

1. ctrl+alt+o → ctrl+alt+a (or /connect) → pick the provider.
2. Paste your API key (or sign in, where the provider offers it).
3. Pick a model.

Your own model server

Run your own model on this PC or another machine on your network? llama.cpp, vLLM, LM Studio, Ollama and any OpenAI-compatible server work.

1. ctrl+alt+o → ctrl+alt+a → Local or network server (first in the list).
2. Type the address, for example 192.168.1.50:8080. http:// and /v1 are added for you.
3. Give it a name, like My GPU server.
4. Type its API key if you started the server with one; otherwise leave it empty.

Orynr Code asks the server what it's serving before saving anything, and tells you plainly if the address, port or key is wrong. It then lists the model that's really loaded, with its size and context length. Adding the same address again updates it.

Think mode

For local models that support it, press ctrl+alt+v (or /variants) to switch between normal, think and think harder. The model's reasoning streams live in a collapsible Thinking block, then the answer.

Orynr Code gives the model room to think: up to 4,096 tokens per step in think and 16,384 in think harder, even if your server was started with a smaller thinking limit (a limit that cuts thoughts off mid-sentence makes a model repeat itself).

Pictures & documents

• Documents: the agent reads PDF, Word, Excel and PowerPoint files directly (a scanned PDF is looked at page by page as pictures).
• Pictures: drag one into the terminal or paste it with ctrl+v. A model that can see images looks at it directly.
• A model that can't see images? Add a picture helper in your settings file (%USERPROFILE%\.config\orynr\orynr.jsonc): "vision_model": "provider/model", any model that can see. It describes each picture for your main model. Orynr Code notices on its own when a model refuses pictures, and remembers it.

Projects & folders

Each project lives in its own folder, normally Documents\Orynr_Projects\<name>, with its own chats, settings and memory. You can also work in any existing folder by naming it in a new chat, or with ora C:\path.

A whole drive, the Windows folder, program folders and your home folder itself can never be used as a project.

Sandbox modes

Every project runs in one of three modes. /sandbox shows the current one and switches it; Orynr Code moves the files for you, never overwrites anything, and undoes a move that fails part-way.

• Docker modes need Docker Desktop. New projects start in Docker mode when Docker is running, and in Local mode otherwise.
• Copy a folder in: from /sandbox, pick any folder on your PC to copy into the project. The AI itself can never reach your other folders.
• In Docker modes nothing from the project runs on your PC: background jobs and the project's checks run inside the container, and tools that would act on your PC (system or network monitor, smart home, the browser, Blender) politely refuse there. Code intelligence and the app preview are off; build, test and run inside the container instead.

For quick snippets there's also a separate safe code runner (Python, JavaScript, Ruby, PHP, Lua) that can't touch your machine at all.

Built-in tools

These come with Orynr Code: always on, nothing to install or start. The agent picks them by itself when a job calls for one. /mcps lists them under Built into Orynr.

With hundreds of tools available, the model only sees a one-line list and loads a tool's details when it needs them, so it stays fast.

Knowledge base

Put your own documents in Documents\Orynr_Knowledge: notes, specs, manuals, PDFs, Word, Excel, PowerPoint, text or code. Orynr Code notices new and changed files by itself and searches them whenever a question might be answered by them, and tells you which file the answer came from.

• /knowledge shows where the folder is and what's in it, and opens it.
• Your files are never changed. Orynr's own notes, journal and search index are kept encrypted.

Saved keys

Some built-in tools need an address, a user name or a password: an MQTT broker, Home Assistant, a database, a web API. Save them with /keys → Save a key, under the name the tool asks for (for example MQTT_HOST, HASS_TOKEN).

• They're stored encrypted. The AI only ever sees the name, never the value.
• When a tool needs one that's missing, the agent tells you the exact name to save. Never paste a password into the chat.
• Database and web API tools use them as vault:NAME.

Blender & Unreal Engine

Blender

1. Switch on blender in /mcps (it's off until you need it), then start a new chat.
2. Ask the agent to “install the Blender add-on”. It puts the Orynr Code bridge into every Blender version on your PC.
3. In Blender: Edit → Preferences → Add-ons, search Orynr Code bridge and tick it. It connects by itself; the Orynr tab in the 3D view's side panel shows the status.

The agent can then read the scene and objects, run Blender Python to build, change, animate or render, and take a clean picture of the 3D view to check its work. The bridge only listens on your own PC and needs a private key that only Orynr Code has; it uses no outside services.

Unreal Engine

Switch on unreal-engine in /mcps, and make sure Unreal Engine is running with its MCP service enabled. Orynr Code then loads its Unreal skills (Blueprints, materials, animation, levels and more) and works in the editor directly. It looks up version-specific details in the documentation instead of guessing.

MCP tool servers

MCP servers give the agents new tools: a database, GitHub, your company's API. Once connected, every agent can use them.

1. Open the MCP screen: /mcps → + Add a server.
2. Choose A web address (remote server) or A program on this PC (local server).
3. Remote: paste the address, often ending in /mcp or /sse. Local: type what starts it, like npx -y @modelcontextprotocol/server-github, or the path to a server file. Paths with spaces are fine.
4. Give it a name, and a key if it needs one: for a web address, a token (sent as Authorization: Bearer, or type Header-Name: value for another header); for a program, NAME=value, for example GITHUB_TOKEN=….

Orynr Code connects and lists the server's tools before saving it, and says plainly what went wrong if it can't. Keys are stored encrypted, never in a settings file. A server on your own PC or network needs one extra “Allow it, it's mine”.

Skills, plugins & memory

• Project memory. The agents save what they learn about a project (decisions, fixes, conventions) and a map of the codebase, and each new session starts with a short since last time brief, including anything left open on the last chat's to-do list. Both are stored encrypted in the project's .orynr folder. /memory shows it; /remember adds to it.
• Project instructions. Put guidance for the agents in ORYNR.md (or CLAUDE.md) in your project. /init writes one for you by analysing the project.
• Skills (/skills) add know-how. Orynr Code uses them by itself when a request matches one, as guidance rather than orders.
• Plugins can add tools, hooks and permission rules. Custom agents can be added in settings, or created by the Agent itself.
• Trust. A project's own tools, skills and settings only load once you trust the project (/trust).

How Orynr Code protects you

Most of this is simply how it's built.

Privacy & your data

• Runs on your computer. Your projects, chats and settings stay on your machine.
• No telemetry, no analytics, no account. Orynr Code sends nothing to Orynr.
• Data leaves only where you send it. Your messages, and the files needed to answer them, go to the model provider you chose. A web search, a page it fetches for you, or a tool server you connected get what they need. A server you run yourself keeps everything in-house.
• Your work is yours. You own the code you create with Orynr Code.

Troubleshooting

Fixes for the things people actually hit. Still stuck? Ask the Agent to diagnose it.

Frequently asked questions