1Password for AI agents
1Password is a password manager and secure vault for storing credentials, notes and shared secrets. Individuals and teams use it to keep logins organized and access controlled. 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 1Password.
Look up a vault item
Retrieve the details of a stored credential or secure note from a specific vault when a workflow needs it.
Update a stored item
Change the fields on an existing password or secure note entry, recorded as a write action.
Delete an item from a vault
Remove a credential or note from a vault, a destructive step that should require explicit approval first.
6 1Password actions, graded by risk.
Every 1Password action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
3Look things up. Allowed by default.
- 1password.get_itemGet Item
- 1password.list_itemsList Items
- 1password.list_vaultsList Vaults
write
2Create and change things. Allow, or ask the user first.
- 1password.create_itemCreate Item
- 1password.update_itemUpdate Item
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- 1password.delete_itemDelete Item
1Password in three steps.
- 01Your users connect 1PasswordThey add their 1Password api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Item” can wait for the user, and “delete Item” can be denied outright.
- 03Any agent can actYour agent calls 1Password through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('1password', { read: 'allow', write: 'ask', // create_item destructive: 'deny', // delete_item }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How 1Password connects.
Users add their 1Password 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 1Password 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
Use 1Password from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraphMore security and identity apps.
1Password and Arc0, answered.
Can I use 1Password with Claude, ChatGPT or Cursor?
Yes. Connect 1Password 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 1Password actions you allow.
How do users connect 1Password?
Users add their 1Password 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.
Which 1Password actions can my agent take?
6 in total: 3 read, 2 write and 1 destructive, such as “create Item”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in 1Password?
Yes. Actions like “delete Item” 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 1Password too?
Yes. The same 1Password connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug 1Password into your agent.
Your users connect 1Password once, under your brand. Your agent gets 6 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan