Do I need to know how to write prompts?

No. That's the whole point of Agent: you describe the idea in plain words, and it composes the professional prompt for the right model itself — with all the structure, lighting, and details you'd otherwise tune by hand. If you do want the prompt as text, ask for "just the prompt" — but that's optional.

What if I don't have a clear idea yet?

Then talk it through in the chat. Tell Agent roughly what you want — it offers a few concrete scene options and helps build a shot list and mood. When the idea is ready, ask it to make it: Agent turns the conversation into a prompt and a generation, without switching to another window.

Is this the same as "Help with the prompt" in the generation bar?

No. "Help with the prompt" improves a prompt for the model and mode you've already picked. Agent is a separate chat: it picks the model itself, explains why, writes the prompt, and — with your approval — runs the generation, or just helps talk through an idea with no generation at all.

Do I get charged tokens just for asking for advice?

Some messages each month aren't charged; beyond that, every turn costs a small flat amount that depends on which Agent model you picked. Generation itself is only charged once you approve a specific proposal with a shown price.

Can I send Agent a photo?

Yes. Attach an image right in the chat — as a reference or for editing; you can brush-select the exact area to change and ask Agent to redo only that. Materials already uploaded to the project are available through the paperclip as well.

Does Agent only run on Claude models?

No. The picker has ten models: Fable 5, Opus 5, Opus 4.8, Sonnet 5, Haiku 4.5, GPT-5.5, GPT-5.6 Sol, Terra, and Luna, and Grok 4.5. GPT-5.6 Terra is the default.

What happens if a generation fails?

Agent shows the error right in the chat, explains the cause, and offers "Fix & retry" — it rewrites the prompt and proposes a new run instead of a blind retry.

Can Agent generate music, voiceover, or avatars?

Music and voiceover — yes: Agent proposes them through the same flow as images and video, with a price card and your "Approve" before launch. Voiceover reads your text verbatim. Avatars still live in their own Studio.

Where are my chats with Agent stored?

Each project has its own chat list, with search and renaming. You can run several chats per project in parallel and switch between them.

July 1, 2026

Agent: you don't need to know how to write prompts — just say what you want

The prompt used to be the real barrier: try describing a scene so the model actually gets it. Agent removes that barrier entirely. Describe your idea in plain words — or talk the scenario through in the chat, like with a colleague — and Agent writes the professional prompt for the right model itself. All you do is Approve: it also picks the model, shows the exact price, and runs the generation.

The hardest part of generation was always the prompt. A good result never came from just "draw a cat" — it took a carefully tuned description: the right structure for the specific model, lighting, camera movement, timecodes for multi-shot clips. Agent takes all of that off your plate. Tell it what you want in plain words, and it composes the professional prompt for the right model itself.

Better still — your idea doesn't even have to be finished. Talk the scenario through right in the chat: sketch out the concept, the shots, the mood together — like with a fellow creator. Agent suggests, pushes back, offers options. When the idea is ready, just ask it to make it — and it turns the conversation into a prompt and a generation.

From there Agent picks a fitting model from the live catalog and shows a proposal card: the model, settings, and the exact cost in tokens — before anything runs. Generation only starts after your Approve. The whole path stays visible in the chat: a "Generating…" status, the finished result previewed right there — with the asset already sitting in your project's Materials.

Agent remembers your project: describe its style, brand, characters, or rules once, and it keeps them in mind in every chat, so you never re-explain the context. And if a generation fails, it won't leave you with a dry "didn't work": it diagnoses the cause, rewrites the prompt, and proposes a new run. You can also run as many chats per project as you need — in parallel, with search and renaming.

On this page

Highlights No prompt to write Price upfront Launch and errors Modes and models Memory and chats Coming soon FAQ

What changed

It writes the prompt

You don't need to know how to write prompts or find the right wording. Say the idea in words — Agent composes the professional prompt for the right model for you.

Talk the idea through

No finished scenario? Build it together right in the chat — the concept, the shots, the mood. When it's ready, ask it to make it.

It picks the model too

Agent chooses a fitting model for the task — no need to figure out which is best for faces, which for motion, which for in-frame text.

Price before you run

Before generating you see a card with the model, settings, and the exact cost in tokens. You're only charged after you hit Approve.

The whole path is visible

Launch, generation status, the finished result, or an error — right in the chat; the result lands straight in your project's Materials.

Fixes on failure

If a generation fails, Agent diagnoses the cause, rewrites the prompt, and proposes a new run — not a blind retry.

Remembers your project

Save style, brand, characters, and rules to your project's memory once — Agent applies them on every turn.

No prompts

The prompt is Agent's job, not yours

Prompting was its own craft for a long time: to get a good frame, "make a clip" isn't enough — you had to know how to describe the scene for a specific model, where lighting matters, where camera movement does, where timecodes for shot changes go. Agent takes that craft on itself. You say the idea in human words — it composes a tuned, professional prompt: for the right model, with the right structure and the details that usually separate "meh" from "wow."

And you don't even have to know exactly what you want. Open the Agent tab in a project and just start talking: "I want a short clip about a coffee shop, but I don't know where to start." Agent offers a few concrete scene options, argues the details, helps build a shot list — and then turns that same conversation into a prompt and a generation. Discussing the idea and making it happen live in one window.

While Agent is working on a reply, you can see exactly what it's doing right now — not a summary after the fact, but step by step: "Thinking through the request," "Searching materials," "Picking a model," "Composing a prompt" — all the way to the finished result.

And if you do want the prompt as text?

Ask for "just the prompt, I'll run it myself" — Agent returns ready text you can copy and paste anywhere, with no generation run and no charge for it. The craft is available by hand too — it's just no longer required.

Proposal

A proposal card: model, settings, and price — before you run it

When Agent is ready to propose a generation, a PROPOSAL card appears in the chat: the model and key settings, a prompt excerpt (expandable to the full text), and the exact cost in tokens — not an estimate range. If you have a Membership discount, the card shows the strikethrough base price next to the discounted total.

From there it's Approve or Reject. Nothing runs and nothing is charged until you hit Approve — even though Agent already spent time picking the model and drafting the prompt for that turn.

The price is protected from drift

The card shows the final price for the current proposal. If the proposal goes stale or the price drifts noticeably by approval time, Agent asks you to refresh the request instead of charging a different amount.

Lifecycle

From launch to result — without leaving the chat

After you approve, the generation runs the same way it would from the editor: the same queue, the same billing, the same engine. Meanwhile the chat shows a live status — "Generating…" — and once the asset is ready, an inline preview appears with a link straight to Materials.

If a provider rejects the generation or something breaks, Agent doesn't just say it failed. It offers "Fix & retry": it works out what went wrong, rewrites the prompt (removing whatever might have triggered a block, or simplifying the request), and proposes a fresh card — with a new approval, not an automatic retry on your dime.

Control

Your choice of approval mode and Agent model

By default Agent runs in Confirm mode: every launch is a card with a price and an explicit Approve. If pausing on every detail isn't what you want, switch to Auto (runs without pausing, up to a daily budget) or Hybrid (auto-runs cheap generations, still shows a card for expensive ones).

You can also pick which model drives the conversation itself — Agent isn't just Claude. The picker has ten options: Fable 5 for maximum quality on the hardest tasks; Opus 5 and Opus 4.8 as two independent versions for complex briefs and fine edits; Sonnet 5 as the versatile Claude; Haiku 4.5 for quick, simple turns; GPT-5.5 for strong reasoning; GPT-5.6 Sol, Terra, and Luna as the flagship, balanced, and fast levels of one family; and Grok 4.5 for bold ideas and an unconventional point of view. Every per-turn price is shown before you choose.

Memory

Agent remembers your project — and you can run several chats

In "Project context" you can save notes by category — Rules, Brand, Personality, Style, Habits, General. For example: "the hero is a ginger cat named Mars" or "no neon." Agent factors these notes into every turn, so you don't have to re-explain context each time.

You can keep more than one chat per project — "New chat" starts another thread with its own title (auto-generated from the first message, but renameable). The chat list is searchable right there on the left, and a link to a specific chat is preserved in the URL.

What's next

What's coming next for Agent

Part of this plan has already shipped: photos can be attached right in the chat (with brush-selected areas for targeted edits), Agent generates music and voiceover, and projects now have employees — eight specialists across content, strategy, and storytelling. See the "Agent 2.0: your project now has employees" announcement for details.

Still on the roadmap:

Skill management

Turn Agent's knowledge packs on and off by hand, for a specific task.

Employees working together

Specialists will learn to hand tasks to each other — from a script all the way to a finished clip with a soundtrack.

This is a plan, not a date

Timing isn't fixed. We'll cover each step separately — keep an eye on the Updates section.

Questions

Frequently asked questions about Agent

Do I need to know how to write prompts?: No. That's the whole point of Agent: you describe the idea in plain words, and it composes the professional prompt for the right model itself — with all the structure, lighting, and details you'd otherwise tune by hand. If you do want the prompt as text, ask for "just the prompt" — but that's optional.
What if I don't have a clear idea yet?: Then talk it through in the chat. Tell Agent roughly what you want — it offers a few concrete scene options and helps build a shot list and mood. When the idea is ready, ask it to make it: Agent turns the conversation into a prompt and a generation, without switching to another window.
Is this the same as "Help with the prompt" in the generation bar?: No. "Help with the prompt" improves a prompt for the model and mode you've already picked. Agent is a separate chat: it picks the model itself, explains why, writes the prompt, and — with your approval — runs the generation, or just helps talk through an idea with no generation at all.
Do I get charged tokens just for asking for advice?: Some messages each month aren't charged; beyond that, every turn costs a small flat amount that depends on which Agent model you picked. Generation itself is only charged once you approve a specific proposal with a shown price.
Can I send Agent a photo?: Yes. Attach an image right in the chat — as a reference or for editing; you can brush-select the exact area to change and ask Agent to redo only that. Materials already uploaded to the project are available through the paperclip as well.
Does Agent only run on Claude models?: No. The picker has ten models: Fable 5, Opus 5, Opus 4.8, Sonnet 5, Haiku 4.5, GPT-5.5, GPT-5.6 Sol, Terra, and Luna, and Grok 4.5. GPT-5.6 Terra is the default.
What happens if a generation fails?: Agent shows the error right in the chat, explains the cause, and offers "Fix & retry" — it rewrites the prompt and proposes a new run instead of a blind retry.
Can Agent generate music, voiceover, or avatars?: Music and voiceover — yes: Agent proposes them through the same flow as images and video, with a price card and your "Approve" before launch. Voiceover reads your text verbatim. Avatars still live in their own Studio.
Where are my chats with Agent stored?: Each project has its own chat list, with search and renaming. You can run several chats per project in parallel and switch between them.