Carbone for AI agents

Carbone generates documents from JSON data and templates, turning structured data into PDFs, Word files, or spreadsheets without manual formatting. 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.

DocumentsAPI keyMCP + RESTcarbone.io
AUDIT LOG · CARBONEPOLICY: acme-support
09:41:07 · claude · u_8f2read
carbone.list_templates
List Templates✓ allowed · 212ms
09:41:08 · claude · u_8f2read
carbone.get_status
Get Carbone Server Status✓ allowed · 164ms
09:41:09 · claude · u_8f2write
carbone.update_template
Update Template Metadata✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
carbone.delete_template
Delete Carbone Template✕ blocked · policy: deny
EVERY CARBONE CALL, ON THE RECORD
01 · USE CASES

What agents do in Carbone.

01

Render a report from a template

Render a template with a set of JSON data to produce a finished PDF or DOCX report.

02

Check available templates

List available templates before choosing which one to use for a new report.

03

Confirm before deleting a template

Confirm before deleting a template, since any report still using it would stop working.

02 · ACTIONS

11 Carbone actions, graded by risk.

Every Carbone action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

6

Look things up. Allowed by default.

  • carbone.get_status
    Get Carbone Server Status
  • carbone.list_templates
    List Templates
  • carbone.list_template_tags
    List Template Tags
  • carbone.list_template_categories
    List Template Categories
  • carbone.generate_report
    Generate Carbone Report
  • carbone.download_template
    Download Template

write

4

Create and change things. Allow, or ask the user first.

  • carbone.update_template
    Update Template Metadata
  • carbone.set_api_version
    Set Carbone API Version
  • carbone.upload_template
    Upload a template
  • carbone.render_template_direct
    Render Template Direct

destructive

1

Delete, cancel or archive. Ask first, or deny outright.

  • carbone.delete_template
    Delete Carbone Template
03 · HOW IT WORKS

Carbone in three steps.

  1. 01Your users connect CarboneThey add their Carbone api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, writes like “update Template Metadata” can wait for the user, and “delete Carbone Template” can be denied outright.
  3. 03Any agent can actYour agent calls Carbone through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('carbone', {
  read: 'allow',
  write: 'ask',        // update_template
  destructive: 'deny',  // delete_template
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How Carbone connects.

Users add their Carbone api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

The same Carbone 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
CREDENTIALS
Per-tenant encrypted vault
MODEL SEES
Results only, never credentials
AUDIT LOG
Every call, on every plan
05 · WORKS WITH

Use Carbone from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
07 · FAQ

Carbone and Arc0, answered.

Q01

Can I use Carbone with Claude, ChatGPT or Cursor?

Yes. Connect Carbone 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 Carbone actions you allow.

Q02

How do users connect Carbone?

Users add their Carbone api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

Q03

Which Carbone actions can my agent take?

11 in total: 6 read, 4 write and 1 destructive, such as “update Template Metadata”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Carbone?

Yes. Actions like “delete Carbone Template” 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.

Q05

Can my own backend call Carbone too?

Yes. The same Carbone connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Carbone into your agent.

Your users connect Carbone once, under your brand. Your agent gets 11 actions behind your policies, with every call on the record.

Free to build · MCP + REST · Audit log on every plan