Product
Hand your agent a scoped key to your brand.
Codexroom runs a Model Context Protocol server inside your workspace. Your agent connects through OAuth with the scopes you pick. It reads the brand, composes the deck, and returns a watermarked link. Your brand team keeps the keys.
In the product
The agent has the language. Codexroom is the library.
Compose over MCP
Prompt to watermarked link, in your brand.
The agent reads the brand, picks a deck template, fills the slots from your data room, validates the result on brand, and returns a share link — all through typed MCP tools. It never writes a slide in raw HTML.
Read the MCP referenceBuild the Series B board deck from our Q3 numbers.
On it — pulling your Q3 metrics and composing on the board template.
Series B — Board Deck
14 slides · ready to share
Scoped, audited access
Your brand team keeps the keys.
The agent borrows a scoped OAuth grant — read the brand, write decks and shares, nothing more — bound to one workspace, one user, one client. Every tool call writes to the audit log under the human who authorised it. Revoke the grant and the agent goes dark.
How governance works- --token-accent#006DD5
- --token-bg-canvas#0F2440
- --token-font-display"Tiempos Headline"
- --token-radius-card12px
- --token-accent-soft#7FB3E8#5A9BDDdraft v5
d.okafor published tokens v4 · 12 templates re-rendered
audit log · every change versioned, nothing overwritten
The mechanism
A Model Context Protocol server inside every workspace.
OAuth with scopes. Your user authorises a specific MCP client (Claude Desktop, Cursor, your internal agent stack) with a specific set of scopes. The common grant is tokens:read, assets:read, templates:read, decks:write, shares:write. Brand-builder agents add tokens:write and templates:write to author the governance layer itself.
Typed tools. The agent sees a list of typed tools that map to Codexroom's API — list_slide_templates, search_assets, create_deck, validate_deck, create_share. Each returns a typed result.
Typed resources. brand://current, templates://slides, data-room://summary. Read-once context the agent loads at the start of a conversation.
Audit on every action. Every tool call writes to the audit log under the user who authorised the grant. Compliance can trace any deck back to the human who said yes.
Why MCP
Your agent stays where it lives.
Codexroom does not have a chat box for your agent to live in. Your team already has one — Claude Desktop, Cursor, the internal copilot the engineering team built last quarter. MCP is the open standard that lets any of them connect to Codexroom on equal footing. Claude Desktop, Cursor, and the OpenAI Agents SDK are tested on day one; the full list and the known quirks live on the developers page.
Reference. The MCP specification lives at modelcontextprotocol.io. Codexroom uses the official @modelcontextprotocol/sdk Node package on the server side.
See it work