# Raster — full context for agents and LLMs > The agent-native DAM for teams that ship. Raster helps humans and AI agents store, organize, search, and ship visual assets. This file concatenates every public Raster marketing page as Markdown, so an agent can load the whole site in one fetch. The complete API reference — REST, GraphQL, MCP, CLI, and developer guides — is a separate one-file dump: https://raster.app/docs/llms-full.txt ## How Raster works (orientation for agents) Raster stores original image files, organizes them with AI-assisted tagging, and serves them from a global CDN — for humans and for AI agents acting on their behalf. The data model is three levels: an **organization** owns one or more **libraries**, and each library holds **assets** (images). An API key is scoped to a single organization and an allowlist of its libraries, each at read or write access, so every asset-level call takes an `organizationId` + `libraryId` pair. One data model and one API key drive three transports — pick the one that fits your client: - **MCP** — connect an agent client (Claude, Cursor, VS Code, the OpenAI Responses API) to the remote server at `https://mcp.raster.app/` with a Bearer key. Reference: https://raster.app/docs/api/mcp - **REST** — call `https://api.raster.app/` over plain HTTP with an `Authorization: Bearer` key and the `Api-Version` header; responses use a `{ data }` / `{ error }` envelope. Reference: https://raster.app/docs/api/rest - **GraphQL** — the same surface at `https://api.raster.app/`. Reference: https://raster.app/docs/api/graphql The working loop is the same across all three transports: 1. **Resolve scope** — `whoami` (REST `GET /me`) returns your `organizationId` and the libraries the key can reach. 2. **Read what exists** — list and search assets and tags to ground later actions. 3. **Add images** — upload from a public `http(s)` URL or an inline base64 payload; processing is asynchronous, and each asset returns a permanent CDN `url` right away. 4. **Organize** — tag, describe, transfer between libraries, or soft-delete (recoverable from trash). 5. **Hand off** — return each asset's permanent CDN `url` and the library URL to the user. An agent with no account and no key can start from zero: one call mints an organization, a starter library, and a key from an email, then returns a claim link the user follows to keep the work. Full per-transport runbook: https://raster.app/docs/guides/ai-agents ## Developer and API resources - [Full API & developer docs](https://raster.app/docs/llms-full.txt): The complete REST, GraphQL, MCP, CLI, and guide reference as one Markdown file. - [OpenAPI spec](https://api.raster.app/openapi.json): Machine-readable OpenAPI 3.1 description of the REST API. - [Using Raster from an AI agent](https://raster.app/docs/guides/ai-agents): Runbook for LLMs — discover Raster, authenticate, and run the upload, tag, search, and hand-off loop. - [Start without an account](https://raster.app/docs/guides/start-without-an-account): One call creates a library and key from an email, so an agent can upload before the user signs up; the user claims it later. --- # The agent-native DAM for teams that ship (https://raster.app) Raster gives humans and AI agents one place to store, understand, retrieve, transform, and deliver visual assets. Fast enough for startups. Structured enough for serious teams. [Get started](/sign-in) · [Explore product](/product) ## Trusted by amazing teams Raster is used by growing teams around the globe. ## Mission AI made image creation instant. Asset management hasn't kept up. Teams now generate, edit, remix, and ship more images than legacy DAMs were designed to handle. The volume changed. The workflow changed. The asset layer has to change too. ## Photo management Raster combines fast libraries, AI-driven organization, nondestructive image work, collaboration, integrations, API access, and delivery in one modern photo manager. ## A million and one ways to use Raster Raster works the way you do, for your team. Here are some popular use cases. - Share wedding photos with family - Make a shared repo for a marketing campaign - Organize every product photo before a launch - Give an agent approved brand imagery - Build a visual library for a client portal - Collect event photos from the whole team - Keep ecommerce variants ready to publish - Store AI generations next to final selects - Hand off design references without a Drive folder - Ship CMS images from one maintained library ## Raster vs legacy DAMs Legacy DAMs were built around admin taxonomies, locked-down portals, manual tagging, and enterprise rollouts. Raster is built around fast libraries, visual context, structured metadata, API-connected workflows, and teams that ship. ## Integrations Raster connects to Contentful, DatoCMS, Figma, Sanity, API workflows, and CDN delivery. --- # The Raster Journal (https://raster.app/blog) ### [A rebuilt Raster](/blog/a-rebuilt-raster) May 15, 2026 — Raster A new foundation for Raster — and the start of a much bigger road ahead. --- ### [Introducing Raster's AI image description feature](/blog/autogenerated-image-descriptions) June 8, 2023 — Edward Hewitt Introduction Raster's new AI-powered feature that auto-generates image descriptions. --- ### [4 reasons why your DAM is more important than you may think it is](/blog/4-reasons-why-your-dam-is-more-important-than-you-may-think-it-is) May 11, 2023 — Edward Hewitt 4 reasons why your DAM should be at the heart of your marketing tech stack. --- ### [Digital asset management and the AI revolution](/blog/digital-asset-management-and-the-ai-revolution) April 12, 2023 — Edward Hewitt Artificial Intelligence is impacting every area of technology, including digital asset management and Raster's AI-powered features. --- ### [Introducing the Raster plugin for Figma ](/blog/introducing-the-raster-plugin-for-figma) February 24, 2023 — Edward Hewitt The Raster plugin for Figma will allow you to use assets from your Raster library directly in Figma. --- ### [How good digital asset management (DAM) can help you to scale your content marketing](/blog/how-digital-asset-management-can-help-you-scale-your-content-marketing-strategy) February 7, 2023 — Edward Hewitt Digital asset management is an often overlooked aspect of marketing strategy, but the right approach can help you scale your content strategy and increase your ROI. --- ### [Best Digital Asset Management software in 2023](/blog/best-dam-software-2023) January 14, 2023 — Edward Hewitt Find out the best Digital Asset Management software options for 2023. --- ### [Why we created Raster](/blog/why-raster) January 14, 2023 — TJ Kohli We created Raster to improve digital asset management for all teams. --- # We're hiring. (https://raster.app/careers) Work at Raster, and build the future of photography management and editing. ## Open positions - **Senior Software Engineer** — [Apply](https://read.cv/teams/raster/senior-sveltekit-developer-ff92WeSwxJAcie7prLDt) - **Senior Design Engineer** — [Apply](mailto:careers@raster.app) - **Junior Design Engineer** — [Apply](mailto:careers@raster.app) - **QA Engineer** — [Apply](mailto:careers@raster.app) - **Prompt Engineer** — [Apply](mailto:careers@raster.app) ## Don't see your superpower? We're always hiring talented, passionate people who want to grow, learn, and kick @$$ together. Join the team. [Send us an email](mailto:careers@raster.app) --- # Raster Changelog (https://raster.app/changelog) - [July 31, 2026](/changelog/2026-07-31) - [July 24, 2026](/changelog/2026-07-24) - [July 17, 2026](/changelog/2026-07-17) - [July 10, 2026](/changelog/2026-07-10) - [July 3, 2026](/changelog/2026-07-03) - [June 26, 2026](/changelog/2026-06-26) - [June 19, 2026](/changelog/2026-06-19) - [June 12, 2026](/changelog/2026-06-12) - [June 5, 2026](/changelog/2026-06-05) - [May 29, 2026](/changelog/2026-05-29) - [May 25, 2026](/changelog/2026-05-25) - [May 15, 2026](/changelog/2026-05-15) - [April 30, 2026](/changelog/2026-04-30) - [April 16, 2026](/changelog/2026-04-16) - [October 30, 2025](/changelog/2025-10-30) - [June 13, 2025](/changelog/2025-06-13) - [May 4, 2025](/changelog/2025-05-04) - [February 17, 2025](/changelog/api-0-3-2) - [February 3, 2025](/changelog/2025-02-03) - [December 1, 2024](/changelog/2024-12-01) - [September 6, 2024](/changelog/contentful-app) - [August 30, 2024](/changelog/datocms-plugin) - [August 13, 2024](/changelog/api-0-2-0-alpha) - [July 4, 2024](/changelog/2024-07-04) - [March 7, 2024](/changelog/2024-03-07) - [August 1, 2023](/changelog/2023-08-01) - [April 1, 2023](/changelog/2023-04-01) - [January 26, 2023](/changelog/2023-01-28) - [December 28, 2022](/changelog/2022-12-28) - [November 8, 2022](/changelog/2022-11-08) - [September 20, 2022](/changelog/2022-09-20) - [September 9, 2022](/changelog/2022-09-09) - [August 16, 2022](/changelog/2022-08-16) - [August 5, 2022](/changelog/2022-08-05) - [July 25, 2022](/changelog/2022-07-25) - [June 31, 2022](/changelog/2022-06-31) - [June 4, 2022](/changelog/2022-06-04) --- # Raster Design System (https://raster.app/design) The open-source Raster Design System. This is our design playground. It is by no means complete, but we welcome you to gain inspiration from it, copy it, and share it with the world. ## Components - Accordion - Avatar - Button - Tag - Color - Form - Tabs - Glass - Icon - Tooltip ## Tabs Two styles from one class. `.tabs` is the default segmented control — the primary tab group, with a neutral pill sliding behind the active tab. `.tabs.tabs-minimal` is the compact underline style, for sub-tabs or where space is tight (a full-contrast label over a raster-red underline). Put the class on whichever element fits semantically: a group of buttons (in-page switching), a group of links (`aria-current="page"` marks the active one), a radio group, or a `