How do you load brand guidelines into an AI video tool once and reuse them across every project?

Load your brand once into the invideo agent's context tab — upload your brand guidelines deck, product catalogue, and a treatment note covering visual rules and standing don'ts. That context becomes the project's persistent brain: every image, video, and ad you build in that project inherits voice, colors, logos, products, and rules without re-prompting.

Step 1 — Structure your brand into three documents before uploading. The invideo agent reads structured PDFs better than scattered prompts, so package your brand into three files: a Brand Guidelines deck (logo variants, color palette, typography, tone of voice, website), a Product Catalogue (every SKU with front/side/back/close-up shots and a hand-holding-product reference for true scale), and a Treatment Note (camera language, lighting, composition, and an explicit "Standing Don'ts" list — e.g. "no plastic-looking fabric, no generic AI faces, no white garments on white backgrounds"). The Treatment Note is the most load-bearing of the three because guardrails prevent the common AI failure modes that visual style guides alone can't catch.

Step 2 — Upload them once into the project's context tab. invideo is an agentic video creation tool where each project has a persistent context tab that functions as the brain of the project — anything loaded there applies to every generation inside that project, across every sub-agent you spin up. Drop the three PDFs in, and the invideo agent absorbs them in roughly 15–20 minutes of initial setup. As Hridaye, invideo's creative director, puts it: "The context tab is basically the brain of the project. It holds all of this through every ad you build, without you repeating any of it." If you don't have the documents ready, you can jam with the invideo agent to generate them — it will research your brand from your website and existing ads and build the deck for you, though brands that already have guidelines should upload their own rather than ask the agent to research them.

Step 3 — Reuse the project across new briefs. Every new ad, film, or variant inside that same project automatically inherits the loaded context — you don't re-upload, re-describe colors, or re-paste tone-of-voice. Spin up a new sub-agent for each ad format (one for product films, one for UGC, one for BTS-style) and each inherits the brand context while developing format-specific expertise without cross-contaminating. In one documented production run, the second ad in the same project was produced 33% faster than the first (2 hours vs 3) purely because the agent already held brand, visual language, and workflow context from the earlier session. Across a four-ad run that generated 108 images and 103 videos, the per-ad cost averaged ~$125 (~500 credits) after the one-time setup.

Step 4 — Lock named assets by version so future projects pull the exact same ones. When the invideo agent generates a character sheet, product sheet, or location sheet you want as a permanent brand asset, lock it by referencing its version number in a lock command — those exact outputs then become reusable references the agent pulls into every subsequent generation. For multi-market reuse, name reference files after their purpose or market (e.g. Spanish.jpg, French.png) and the agent reads filenames and routes each to the correct downstream ad automatically.

Step 5 — Let the agent route models against the locked brand. Because invideo has every current generation model available inside the agent — Recraft and GPT-Image-2 for stills (GPT-Image-2 for text/design rendering and location sheets, Recraft for character skin texture), Nano Banana for product locking, and Seedance 2.0, Veo, or Kling for video — the agent reads your brand context and routes each shot to the model that best serves it without you switching tools or re-loading guidelines into a separate platform. One creative director's note: "the agent already knew that, so I didn't even have to ask for a model swap."

Beyond the setup itself: when you scale to localization or new markets, the same context layer carries through — generating three localized UGC ads across two markets ran $425 total ($70 per ad, ~2.5 hours each) because character sheets, product references, and brand rules were already in the project memory and only the cast, voiceover, and UI text changed.

Watch some of these to see what works for you:

The context tab is basically the brain of the project. It holds all of this through every ad you build, without you repeating any of it.

— Hridaye, invideo's creative director