Zeplin for AI agents
Zeplin connects designers and developers by sharing design specs, assets, and components, used by product teams handing off finished designs for implementation. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Zeplin.
List a screen's components
Pull the components used on a design screen, a read-only handoff check.
Get a project's colors
Read the defined color palette for a project before implementing it.
Confirm before updating a styleguide color
Check with the user before changing a styleguide color, since components across the project may reference it.
24 Zeplin actions, graded by risk.
Every Zeplin action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
0Look things up. Allowed by default.
- No read actions.
write
23Create and change things. Allow, or ask the user first.
- zeplin.projects_getGet Zeplin Project by ID
- zeplin.screen_section_getGet Screen Section
- zeplin.screen_version_getGet Screen Version
- zeplin.project_colors_listList Project Colors
- zeplin.project_color_updateUpdate Project Color
- zeplin.screen_sections_listList Screen Sections
- zeplin.screen_versions_listList Screen Versions
- zeplin.auth_o_auth_authorizeStart OAuth authorization (PKCE)
- zeplin.screen_annotation_getGet Screen Annotation
- zeplin.screen_components_listList Screen Components
- zeplin.screen_versions_createCreate Screen Version
- zeplin.styleguide_colors_listList Styleguide Colors
- zeplin.projects_members_inviteInvite Project Member
- zeplin.screen_annotations_listList Screen Annotations
- zeplin.styleguide_color_createCreate Styleguide Color
- zeplin.styleguide_color_updateUpdate Styleguide Color
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- zeplin.screen_annotation_deleteDelete Screen Annotation
Zeplin in three steps.
- 01Your users connect ZeplinThey sign in to Zeplin on Arc0 Connect, under your brand, and approve the access you ask for.
- 02You set the rulesReads run, writes like “get Zeplin Project by ID” can wait for the user, and “delete Screen Annotation” can be denied outright.
- 03Any agent can actYour agent calls Zeplin through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('zeplin', { read: 'allow', write: 'ask', // projects_get destructive: 'deny', // screen_annotation_delete }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Zeplin connects.
Users sign in to Zeplin on Arc0 Connect and approve the scopes you request. Build with Arc0’s Zeplin OAuth app, or bring your own so the Zeplin consent screen names you. Tokens refresh automatically, and you can export them whenever you want.
The same Zeplin connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key · OAuth 2.0
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Zeplin from any agent.
Zeplin and Arc0, answered.
Can I use Zeplin with Claude, ChatGPT or Cursor?
Yes. Connect Zeplin to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Zeplin actions you allow.
How do users connect Zeplin?
Users sign in to Zeplin on Arc0 Connect and approve the scopes you request. Build with Arc0’s Zeplin OAuth app, or bring your own so the Zeplin consent screen names you. Tokens refresh automatically, and you can export them whenever you want.
Which Zeplin actions can my agent take?
24 in total: 0 read, 23 write and 1 destructive, such as “get Zeplin Project by ID”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Zeplin?
Yes. Actions like “delete Screen Annotation” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Zeplin too?
Yes. The same Zeplin connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Zeplin into your agent.
Your users connect Zeplin once, under your brand. Your agent gets 24 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan