Clientary for AI agents

Clientary manages the client-facing side of a service business, covering proposals, invoices, projects, and staff hours in one connected system. 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.

Connect Clientary See the 57 actions ↓

Project managementAPI keyMCP + RESTclientary.com

AUDIT LOG · CLIENTARYPOLICY: acme-support

09:41:07 · claude · u_8f2read

clientary.list_leads

List Leads✓ allowed · 212ms

09:41:08 · claude · u_8f2read

clientary.get_hour

Get Hour Entry✓ allowed · 164ms

09:41:09 · claude · u_8f2write

clientary.create_lead

Create Lead✓ approved · approved by user

09:41:12 · claude · u_8f2destructive

clientary.delete_hour

Delete Hour Entry✕ blocked · policy: deny

EVERY CLIENTARY CALL, ON THE RECORD

01 · USE CASES

What agents do in Clientary.

Create a client invoice

Create a new invoice for a client and attach it to the right project.

Look up a project's hours

Look up logged hours on a project before including them in a client invoice.

Confirm before deleting a client

Confirm before deleting a client record, since their projects and invoices reference it.

02 · ACTIONS

57 Clientary actions, graded by risk.

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

read

Look things up. Allowed by default.

clientary.get_hour
Get Hour Entry
clientary.get_lead
Get Lead
clientary.get_task
Get Task
clientary.get_staff
Get Staff
clientary.get_client
Get Client
clientary.list_leads
List Leads
clientary.list_staff
List Staff
clientary.list_tasks
List Tasks
clientary.get_contact
Get Contact
clientary.get_expense
Get Expense
clientary.get_invoice
Get Invoice
clientary.get_project
Get Project
clientary.get_estimate
Get Estimate
clientary.list_clients
List Clients
clientary.list_expenses
List Expenses
clientary.list_payments
List Payments

+ 14 MORE

write

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

clientary.create_lead
Create Lead
clientary.update_hour
Update Hour Entry
clientary.update_task
Update Task
clientary.create_task2
Create Task (v2)
clientary.create_client
Create Client
clientary.update_client
Update Client
clientary.create_contact
Create Contact
clientary.create_expense
Create Expense
clientary.create_invoice
Create Invoice
clientary.create_project
Create Project
clientary.update_expense
Update Expense
clientary.update_invoice
Update Invoice
clientary.update_project
Update Project
clientary.create_project_hours
Create Project Hours
clientary.send_invoice_message
Send Invoice Message
clientary.update_clientary_lead
Update Lead

+ 2 MORE

destructive

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

clientary.delete_hour
Delete Hour Entry
clientary.delete_lead
Delete Lead
clientary.delete_client
Delete Client
clientary.delete_payment
Delete Payment
clientary.delete_payment_profile
Delete Payment Profile
clientary.delete_clientary_contact
Delete Contact
clientary.delete_clientary_expense
Delete Expense
clientary.delete_clientary_project
Delete Project
clientary.delete_recurring_schedule
Delete Recurring Schedule

03 · HOW IT WORKS

Clientary in three steps.

01Your users connect ClientaryThey add their Clientary api key on Arc0 Connect, under your brand. It goes straight into the vault.
02You set the rulesReads run, writes like “create Lead” can wait for the user, and “delete Hour Entry” can be denied outright.
03Any agent can actYour agent calls Clientary 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('clientary', {
  read: 'allow',
  write: 'ask',        // create_lead
  destructive: 'deny',  // delete_hour
})

# 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 Clientary connects.

Users add their Clientary 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 Clientary 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 Clientary from any agent.

Claude

ChatGPT

Cursor

Codex

VS Code

OpenAI Agents SDK

Claude Agent SDK

Vercel AI SDK

Mastra

LangGraph

06 · RELATED

More project management apps.

Google Tasks18 actions

Moco218 actions

Wrike144 actions

ALL PROJECT MANAGEMENT APPS →

07 · FAQ

Clientary and Arc0, answered.

Q01

Can I use Clientary with Claude, ChatGPT or Cursor?

Yes. Connect Clientary 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 Clientary actions you allow.

Q02

How do users connect Clientary?

Users add their Clientary 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 Clientary actions can my agent take?

57 in total: 30 read, 18 write and 9 destructive, such as “create Lead”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Clientary?

Yes. Actions like “delete Hour Entry” 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 Clientary too?

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

Get started

Plug Clientary into your agent.

Your users connect Clientary once, under your brand. Your agent gets 57 actions behind your policies, with every call on the record.

Connect Clientary Browse all apps →

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