OnBrand by SlideSpeak

About OnBrand by SlideSpeak

OnBrand is an MCP (Model Context Protocol) server that feeds brand guidelines directly into AI agents. It supplies colors, logos, icons, imagery, and slide layout rules so tools like Claude, Codex, and ChatGPT produce assets aligned with an organization's visual identity. The system can also generate a design.md file from an existing website, brand documentation, or PowerPoint templates.

Review

OnBrand addresses a specific friction point in AI-assisted design: generic, off-brand output. Instead of wrapping another chat interface around a model, it operates as context injected into the agent's workflow. The tool was released this week and is currently free.

Key Features

• MCP-based design context injection for AI agents (Claude, Codex, ChatGPT).
• Automatic generation of a design.md file from an uploaded website, existing guidelines, or PowerPoint templates.
• Validation steps exposed to the model so the agent can check its own output against the brand rules mid-task.
• Open-source code available on GitHub.
• Works with a free SlideSpeak account - no separate subscription for OnBrand itself.

Pricing and Value

OnBrand is free at launch. Users need a SlideSpeak account to access it, but there are no listed charges for the MCP service. The makers haven't announced any paid tiers or future pricing changes.

Pros

• Direct, persistent brand context reduces the likelihood of AI-generated slides drifting away from guidelines.
• Ingestion from multiple source types means teams don't have to manually rewrite their brand rules for the AI.
• Open-source code allows inspection and self-hosting if desired.
• Validation steps give the agent a way to self-correct, not just follow a static prompt.

Cons

• Launched this week, so there's no established track record or broad user feedback yet.
• Requires setting up an MCP-compatible agent and a SlideSpeak account - not a plug-and-play standalone app.
• Not well suited for teams that don't use AI agents or for those needing a complete design generation suite without external tooling dependencies.

Organizations already experimenting with AI agents for content creation will find OnBrand's approach practical, especially if they regularly fight off-brand slide decks and infographics. Teams without an existing AI agent workflow might find the initial integration steps less relevant until they adopt that architecture.