THE AGENTIC CODING EDITOR

THE AGENTIC CODING EDITOR



★ Super modern, extremly flexible UI
★ Full agentic coding loop workflow
★ Workspace and specification driven
★ Local, external or CLI version agents
★ Save memory & history in your projects
★ Organize your projects on your computer
★ File based, rich media/3D support
★ No login, no subscription, no tracking
Your Human/AI-Workspace To Bring Your Projects To Life

Your Tool


Create projects with Rojos - AI agents that understand your workspace, specifications, tasks, knowledge, files and development history.

Edit documents, knowledge and code manually by yourself or with the AI directly on your computer, no cloud needed.

Your Projects


Develop digital products like websites, apps, software, tools, games, experiences or art.

But also use it to write theatre scripts, production schedules, preparations, IP development, wikis, plans, to-do lists and more to organize and support all your projects.
GIMME' A TOUR SHORT PRESENTATION TO GET YOU STARTED I WANNA ASK FAQ/INTERACTIVE TEXT GUIDE I NEED TO TRY LET'S GET REAL AND TRY THE DEMO ONLINE 9.99 + 1 Year Updates No Subscription Buy & Own details & terms

ROJOS


YOUR PERSONALIZED AGENTS



Define your own agents with unique looks, LLMs and initial prompts to pre-program your agents.

Work together on files directly in the editor and save learnings, corrections and memory for each agent individually, locally in the project.

PAGE EDITING


FOR HUMANS AND AGENTS



Edit your documents, speficiations, tasks and notes with a all-in-one rich WSIWYG editor.

Edit pages, that your Rojos created, or give Rojos pages as to-dos or templates.

TOOLS


ALL THE THINGS



Roject comes with a lot of handy tools for your daily work: Code editor, terminal, file tree, static server, web browser, media browser/viewer (audio, video, 3D), search and more.

And more to come: Tools that will interconnect your agent workflows, git in-editor integration, custom programs for your Rojos, own templates and more.


The Roject Guide What is Roject?

A shared workspace for humans and AI

Roject is a workspace built for the human-agent loop: an editor where people and AI agents work on the same project, in the same window, across as many sessions as it takes. It combines a document editor for structured project knowledge, a code/file editor, and an AI chat panel called Rojo Chat into one tool.

Instead of treating "chat with an AI" and "edit my project" as two separate activities, Roject merges them: the workspace you write specs, tasks, and history into is the same workspace the AI reads before it starts working, and the same one it updates when it's done.

What problem is Roject trying to solve?

Context gets lost between sessions

Chatting with an AI about a project is great in the moment, but the moment usually doesn't survive: decisions, structure, and history live in a scrollback nobody reads again. Next session, you're re-explaining the project from scratch, and the AI starts from zero.

Roject solves this by giving every project a persistent, structured workspace - outline, boards, guides, history, reference - that both you and the agent read and write directly. Nothing needs to be re-explained because it's already there, in a form an agent can navigate on its own.

What does "agentic development loop" mean in Roject?

Read, act, record - repeat

The agentic development loop is the cycle an AI agent runs on a real project: read enough of the workspace to understand the task, make the actual change (write code, edit a spec, update a board), then record what happened so the next session - human or agent - can pick up where this one left off.

In Roject that loop isn't a metaphor, it's the literal file structure: outline for direction, boards for active work, history for what happened, reference for how things work. Each pass through the loop leaves the workspace a little richer than it found it.

What is the difference between working with an AI in a chat and working with an AI inside a Roject workspace?

A scrollback vs. a shared brain

A plain AI chat is a conversation: useful advice, code snippets, explanations - all of it living in a transcript that's easy to lose and hard to search. Nothing about the conversation is structured, so nothing about it is reusable except by re-reading it yourself.

A Roject workspace turns the same kind of work into durable, structured artifacts - a task board, an updated spec page, a history entry - that both of you can navigate directly next time, instead of scrolling back through last week's chat to remember what was decided.

Is Roject primarily an IDE, an AI chat, an agent framework, or something else?

All three, deliberately merged - and none of them required

Roject is a bit of each, on purpose. It has a real multi-panel code and file editor like an IDE, a genuine AI chat panel with tool use like a chat app, and configurable AI agents ("Rojos") with their own entry points into the project like an agent framework.

Importantly, it isn't dependent on agents at all - it's a genuinely capable editor for manual work on its own, with a file browser, a media browser, a web browser, and browser-like navigation inside the page editor itself (follow a link to another page, then go back, the way you would in a web browser). None of those pieces work as well alone as they do combined: the point is one place where manual editing, chatting, and agent work all happen against the same shared project state, and you can lean on any mix of them.

What is the basic idea behind organizing a project as a workspace?

A directory of typed knowledge, not a wall of text

A Roject workspace is a workspace/ directory full of .page files - outline, boards, guides, history, reference - each one a structured, interactive document instead of a flat wiki page. Tasks live as real kanban boards, not bullet lists; guides are linked, not buried in one giant file.

Organizing information this way means both a human skimming the project and an agent parsing it can find the right piece without reading everything: the outline gives the big picture, and everything else is one link away when it's actually needed.

Why is project context such an important problem when working with AI?

AI is only as good as what it's allowed to see

An AI that doesn't know your project's history, conventions, or current priorities will confidently produce answers that are technically fine and practically wrong - reinventing a decision you already made, ignoring a constraint that isn't written anywhere it can read.

The fix isn't a bigger context window, it's better-organized context: giving the agent a workspace it can navigate to find exactly the outline, spec, or history entry relevant to the task at hand, instead of hoping everything relevant fits in one prompt.

How does Roject help humans and AI work on the same project without requiring everything to be understood at once?

Hierarchical, on-demand knowledge

Roject's workspace is organized top-down: a short outline gives the big picture, and boards, guides, history, and reference sit one layer below with links out to more detail only when it's relevant. Nobody - human or agent - has to read the entire project before doing anything useful.

An agent starting a task reads the outline and the specific board or guide the task touches, not the whole workspace. A human returning after a month reads the same layers, in the same order, and ends up with the same picture the agent has.

What is a typical workflow when starting a project in Roject?

From blank slate to structured workspace in minutes

Create a project and it starts from a real workspace template, not a blank index.html: an outline, empty boards, a getting-started guide tailored to the kind of project it is - software, game dev, production, research, and more. You (or an agent) fill in the outline first, then start opening tasks on the boards.

From there it's the loop: pick a task, work it with or without an agent, log what happened. The structure is there from minute one, so there's no separate "documentation phase" bolted on at the end.

What makes Roject different from simply giving an AI access to a large codebase or a huge collection of documents?

Structure beats size

Dumping a whole codebase or document pile at an AI trades one problem (no context) for another (too much undifferentiated context) - the agent still has to guess what's relevant, and a raw file tree doesn't tell it what's a decision, what's a task, or what's stale.

Roject's workspace is typed: boards are boards, history is history, guides are guides - not just files that happen to contain words. That structure is what lets an agent (or a person) retrieve the right slice of a large project instead of searching all of it every time.

The Workspace Guide What is a workspace in Roject?

The project's shared memory, as files

A workspace is the workspace/ directory that ships with every Roject project: outline, boards, guides, history, and reference, all stored as clean, readable .page HTML files on disk. It's version-controllable, backupable, and readable outside of Roject entirely - it's just files.

Inside Roject, the same directory is rendered as rich, interactive pages: boards become kanban columns you can drag cards on, guides become linked documents with real formatting. It's a workspace in the literal sense - somewhere both you and an agent go to do the actual work of understanding the project.

Why doesn't Roject treat the entire project as one huge knowledge space?

One flat blob doesn't scale

A single giant document (or one enormous prompt) forces every reader - human or AI - to wade through everything to find the one paragraph that matters. It also makes it nearly impossible to know what's current versus historical, or what's a decision versus a passing idea.

Splitting the workspace into distinct areas - outline for direction, boards for active state, history for what already happened, guides and reference for how things work - means each piece can be read (or skipped) on its own terms, and each one stays a manageable size.

How can a large project be divided into smaller pieces of information?

By role, not by size

Roject divides a workspace by what kind of knowledge something is, not by arbitrary chunking. Boards hold active tasks and bugs. History holds a dated log of what was built and why. Guides hold conventions and how-tos. Reference holds detailed technical documentation for specific systems.

Each area also nests: a guide's index page gives the overview, and links out to detail pages only for the parts that need more depth. A large project ends up as a tree of right-sized documents instead of one document that keeps growing forever.

What kinds of things can exist inside a workspace besides code?

Anything the project needs to remember

Tasks and bugs on kanban boards, specifications and design notes, mockups and reference images, session history, onboarding guides, architecture references, even notes for a theatre production's prop list or a game's enemy roster - the .page format is generic enough to hold structured content for any kind of project, not just software.

Because pages support typed blocks - lists, images, code snippets, headings - a workspace can express a lot more than prose: a task board is a real board, not a bullet list pretending to be one.

How do specifications fit into the workspace?

Specs live where the agent can find them

A specification for a feature, a level, or a scene is just another page in the workspace - usually linked from the outline or a board, so anyone (or anything) starting related work finds it naturally instead of having to be told where it lives.

Because it's a real page, a spec can carry structure the whole way through: a task board next to it tracking implementation status, a history entry recording when it changed and why, images or code blocks illustrating the design directly alongside the prose.

How can an AI focus on one specific part of a project without needing to understand everything else?

Enter through the piece that matters

Because the workspace is hierarchical and linked, an agent doesn't have to load the whole project to work on one part of it - it can be pointed at (or navigate to) the one board, guide, or spec relevant to the current task and stop there.

This is also how a Rojo's own "entry point" works: a project-focused Rojo can start already anchored to a specific reference page, so every session begins scoped to the right part of the workspace instead of the whole thing.

What does it mean to enter a workspace through a specific reference or starting point?

Where you begin shapes what you see

Instead of always starting from the workspace root, a Rojo (or a human) can be configured to start from a specific page - a guide, a board, a reference doc - and treat that as its first stop, following links out from there as needed.

This matters because it changes what "relevant" means by default: a Rojo entering through a game design outline behaves differently from one entering through a backend reference doc, even inside the exact same project, simply because of where it started reading.

How are project history, research, tasks and decisions useful to an AI working on the project?

Precedent an agent can actually read

History entries record what was built, when, and why - including decisions that didn't make it into any spec because they were made in passing. An agent reading history before starting work avoids repeating a debugging dead end or re-litigating a choice that was already settled last week.

Tasks and boards do the same for current state: an agent can see what's already in progress, what's blocked, and what's already done, instead of duplicating work or contradicting an approach already underway elsewhere in the project.

How can information be organized at different levels of detail?

Outline first, detail on demand

Every area of the workspace follows the same shape: a short, compact index page gives the big picture in a few paragraphs, and links out to detail pages for anything that needs more depth - a specific guide, a specific day's history, a specific reference doc.

Readers - human or AI - are expected to read the index layer fully, then follow only the links relevant to what they're doing. Nobody has to read every detail page just to understand what a part of the project is for.

How does the workspace help both humans and AI navigate a complex project without consuming all information linearly?

Navigation, not consumption

Because everything is linked and layered rather than laid out as one long document, navigating the workspace looks more like following references in a codebase than reading a book front to back - jump to the board that matters, open the one guide that's relevant, skip everything else.

That's true for a human skimming a project they haven't touched in months, and it's true for an agent deciding what to read before starting a task - both move through the same structure, just faster or slower depending on how much context they already carry.

The Agentic Loop Guide What is an agentic loop?

Understand, act, record - then do it again

An agentic loop is the repeating cycle an AI agent goes through while actually working on a project, rather than just answering a single question: gather enough context to understand a task, take real action (edit files, update a board, write a spec), then leave a record of what happened.

Roject is built around making that loop sustainable across many sessions, not just one: because each pass writes back into the same structured workspace, the next loop - run by the same agent, a different Rojo, or a human - starts from where the last one actually left off.

How does an AI move from understanding a task to actually changing or developing a project?

From reading to writing

The agent starts by reading whatever part of the workspace the task points to - a board entry, a spec, a guide - to understand what's being asked and what already exists. From there it moves into the multi-panel editor and makes the actual change: editing code, updating a page, restructuring a board.

The key is that "understanding" and "changing" happen in the same environment against the same files, so there's no lossy translation between what the agent read and what it acts on - it edits the very documents it was reading.

What information does an AI need before it can start working on a task?

Enough, not everything

At minimum: the outline for overall direction, the specific task or bug description, and whatever guide or reference doc governs the part of the project it touches. Beyond that, it needs to know if related work already happened - which is what history is for.

Roject's convention is deliberately "read the first layer, then follow links only as needed" - an agent that reads the compact overview pages first rarely needs to read much else before it has enough to start.

How can an AI discover the relevant parts of a workspace while it is working?

Following links, not searching blind

Because pages link to each other - a task references a spec, a spec references a guide, a guide references a reference doc - an agent can discover what it needs by following references outward from wherever it started, the same way a person would click through a wiki.

When following links isn't enough, the workspace is also just a real file tree, so an agent can search or browse it directly - but the linked structure is what usually gets it there faster than a blind search would.

Why is it useful for an AI to retrieve information in smaller, relevant pieces instead of receiving the entire project context?

Precision over volume

Feeding an agent an entire project up front doesn't make it smarter - it makes the truly relevant information harder to find inside all the noise, and it burns context budget on things that have nothing to do with the current task.

Retrieving smaller, targeted pieces - the one board, the one guide, the one history entry that actually matters - keeps the agent's attention on what's relevant and leaves room in context for the actual work: reading code, reasoning about a fix, drafting new content.

How can specifications guide an AI during development?

A spec is a contract the agent can read

A specification written as a workspace page gives an agent something concrete to build against instead of an ambiguous instruction - what a feature should do, what it shouldn't, what the intended structure looks like. That's a much stronger anchor than a one-line task title.

Because the spec lives in the same workspace as the code, the agent can keep both open at once, checking its work against the spec directly rather than relying on memory of a conversation from earlier in the session.

How do tasks and project history become part of the development process?

The loop writes back

Finishing a task isn't just closing it out - it's expected to leave a trace: the board moves to "done," and a history entry records what was built and why, especially any decision that wouldn't otherwise show up anywhere in the code itself.

In practice a session typically ends by running an action like "Update History": it writes the day's history entry, updates whatever reference docs actually changed, and commits (and pushes) the result. That trace is what makes the next loop faster - whether it's the same agent next week or a different one entirely, reading the board and recent history is what lets a new session start already caught up, instead of starting cold.

What happens when an AI needs more information than it currently has?

Follow another link, or ask

Usually it just follows another link - opens the reference doc a guide pointed to, checks a related board, reads further back in history. The workspace is designed so most gaps can be closed by navigating one layer deeper rather than guessing.

When that's not enough - the missing information only exists in someone's head, or requires a judgment call - the expectation is that the agent stops and asks, rather than confidently filling the gap with a guess. Roject's own contributor guides bake this in as a standing rule.

How can the human guide, correct or redirect the AI during an agentic development loop?

Corrections become part of the record

Directly, in the moment - the human can redirect an agent mid-task the same way they would a colleague, by pointing out a misread requirement or a wrong assumption. Roject encourages logging real corrections, not just fixing and moving on, so the same mistake doesn't repeat next session.

Because the workspace is shared, a correction can also be structural: editing the guide or spec the agent misread, so the fix benefits every future session, not just the current conversation.

How does Roject support long-running development where new tasks, research, specifications and history continuously become part of the project?

Built to accumulate, not reset

Every session adds to the same durable workspace instead of starting from scratch: new tasks land on boards, new decisions get written into specs, and each day's work can be logged to history. None of it depends on a chat transcript surviving.

Over weeks or months, the workspace becomes a genuine project memory - richer with every session rather than static - which is exactly the target audience: sustained projects, not one-shot "generate me an app" requests.

The Rojo Guide What is a Rojo?

A pre-programmed agent, not just a chat window

A Rojo is a named, configurable AI agent inside Roject - its own appearance, its own color, its own model, and its own entry point into a workspace (or none at all, for a general-purpose assistant). You talk to it through Rojo Chat, the same way you'd chat with any AI, but everything about how it behaves is something you set up ahead of time: not just what it knows, but how it acts.

You can have several Rojos in one project, each suited to a different kind of work, and switch between them the same way you'd switch tabs.

Is a Rojo simply another name for an AI agent?

An agent, plus configuration and identity

Mostly yes, with the important part being "configurable." A Rojo bundles an underlying AI agent together with a persistent identity - a name, an appearance, a color theme - and a set of settings: which model it uses, where it starts reading in the workspace, what its persistent memory (corrections, guides) looks like.

That bundling matters because it means the same underlying model can behave like several distinct collaborators depending on which Rojo you're talking to, each with its own accumulated context and habits.

Why would I create different Rojos instead of always using the same AI configuration?

Different jobs want different setups

A Rojo scoped to one codebase and anchored to its reference docs will behave very differently - and more usefully, for that codebase - than a generic assistant with no starting point. Creating separate Rojos lets each one specialize instead of asking one configuration to be good at everything.

It also keeps memory clean: a Rojo's corrections, guides, and history are its own, so teaching one Rojo project-specific conventions doesn't clutter or confuse a completely different Rojo working on something else.

What can be configured differently for each Rojo?

Identity, model, starting point - and behavior

Visually: a Rojo has its own appearance (built from a customizable character rig) and its own color, which carries through the whole Rojo Chat interface when you're talking to it. Functionally: its underlying model - a powerful hosted model or a smaller local one - and its entry point into a workspace, meaning where it starts reading and what it treats as relevant by default.

That entry point does more than hand over knowledge, though - it programs behavior too. The same starting prompt or read-up page can teach a Rojo which actions or skills it's allowed to reach for and when, to actively watch for and flag corrections instead of guessing past them, how much thinking or reasoning a task deserves before it acts, and how it should talk to you along the way. Its accumulated memory - corrections, guides, session history - is also per-Rojo, so each one develops its own track record over time.

How does a Rojo's entry point into the workspace affect what it knows and what it can do?

Starting point sets the default lens - and the default behavior

A Rojo anchored to a specific reference page, guide, or project starts every session already oriented - it reads that starting point first and follows links out from there, the same way a new team member would if you handed them one document to start with.

That doesn't wall it off from the rest of the workspace; it can still navigate anywhere a link (or a search) leads. But "what it can do" isn't only about knowledge either - the same entry material can program behavior: which actions it reaches for on its own, whether it treats an unexpected user message as a correction worth logging, and how it's expected to interact with you along the way.

What is the difference between a Rojo that starts with a specific project reference and a completely fresh generic Rojo?

Anchored vs. blank-slate

A project-anchored Rojo begins every session already pointed at that project's outline or a specific reference doc - it behaves like someone who's read the onboarding guide before their first day. A generic Rojo starts with no workspace context at all, closer to a plain AI chat.

Neither is strictly better - a generic Rojo is the right tool for a quick question with no project baggage, while an anchored one is the right tool for sustained work on one specific thing.

When would I create a project-focused Rojo such as "Code Clawt"?

When the same context comes up over and over

Create a project-focused Rojo when you're doing sustained work on one codebase or project and you're tired of re-establishing context every session - a Rojo anchored to that project's reference docs skips that step by design, and accumulates project-specific corrections and habits over time.

It's also useful when a project has real conventions worth enforcing consistently - a project-focused Rojo that's read the guides once tends to keep applying them, session after session, more reliably than re-explaining them each time.

When would I use a generic Rojo such as "Chatty Cathy"?

For anything that isn't "this project"

A generic Rojo is the right call for quick questions, brainstorming, or work that doesn't belong to any one project's workspace - it behaves like a normal AI chat, without the overhead (or the benefit) of being anchored anywhere in particular.

It's also a reasonable default when you're not sure yet whether a topic deserves its own project-focused Rojo - start generic, and create a dedicated, anchored Rojo later once the work has settled into something worth building sustained context around.

Can different Rojos use different models, such as a powerful online model and a small local quantized model?

Yes - model choice is per Rojo

Each Rojo's model is part of its configuration, so nothing stops you from running a powerful hosted model for a Rojo doing serious development work, and a small local model - even a quantized one running entirely on your machine - for a Rojo handling lighter, more routine questions.

Local models can connect through Roject's tunnel.rokojori.com bridge, so a fully local, private Rojo and a cloud-backed one can sit side by side in the same project without either setup interfering with the other.

How should I decide which Rojo configuration is appropriate for a particular task or question?

Match anchoring and power to the job

Ask two things: does this task need project-specific context, and does it need a powerful model to do well? A quick generic question needs neither - reach for a generic Rojo. Sustained work on a specific project benefits from an anchored one that already knows the territory.

Model choice follows the same logic: reserve a powerful (and possibly costlier) model for work that actually needs the reasoning power, and let a lighter local model handle routine, low-stakes questions where speed and privacy matter more than raw capability.