Longbridge for AI agents
Longbridge is a brokerage platform giving investors market data, account balances, and trading access across stocks and other securities from one connected account. 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 Longbridge.
Check account balance
Look up account balance and buying power before deciding on a trade, read-only.
Pull a stock's candlestick data
Fetch recent candlestick price data for a ticker to review a stock's recent movement.
Require approval before placing an order
Hold any order submission for explicit approval, since trades move real money.
161 Longbridge actions, graded by risk.
Every Longbridge action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
1Look things up. Allowed by default.
- longbridge.estimate_max_purchase_quantityEstimate max purchase quantity
write
152Create and change things. Allow, or ask the user first.
- longbridge.create_watchlist_groupCreate watchlist group
- longbridge.update_watchlist_groupUpdate watchlist group
- longbridge.nowNow
- longbridge.newsNews
- longbridge.depthDepth
- longbridge.quoteQuote
- longbridge.topicTopic
- longbridge.tradesTrades
- longbridge.anomalyAnomaly
- longbridge.brokersBrokers
- longbridge.companyCompany
- longbridge.filingsFilings
- longbridge.signalsSignals
- longbridge.dca_listDca list
- longbridge.dca_stopDca stop
- longbridge.depositsDeposits
destructive
8Delete, cancel or archive. Ask first, or deny outright.
- longbridge.cancel_orderCancel order
- longbridge.delete_watchlist_groupDelete watchlist group
- longbridge.grid_cancelGrid cancel
- longbridge.alert_deleteAlert delete
- longbridge.grid_suspendGrid suspend
- longbridge.alert_disableAlert disable
- longbridge.sharelist_deleteSharelist delete
- longbridge.sharelist_removeSharelist remove
Longbridge in three steps.
- 01Your users connect LongbridgeThey sign in to Longbridge on Arc0 Connect, under your brand, and approve the access you ask for.
- 02You set the rulesReads run, writes like “create watchlist group” can wait for the user, and “delete watchlist group” can be denied outright.
- 03Any agent can actYour agent calls Longbridge through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('longbridge', { read: 'allow', write: 'ask', // create_watchlist_group destructive: 'deny', // delete_watchlist_group }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Longbridge connects.
Longbridge runs its own MCP server behind OAuth. Arc0 registers the client, users approve access on Arc0 Connect, and your agent reaches Longbridge through the same endpoint, policies and audit log as every other app.
The same Longbridge 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
- OAuth 2.0 (MCP)
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Longbridge from any agent.
More finance and accounting apps.
Longbridge and Arc0, answered.
Can I use Longbridge with Claude, ChatGPT or Cursor?
Yes. Connect Longbridge 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 Longbridge actions you allow.
How do users connect Longbridge?
Longbridge runs its own MCP server behind OAuth. Arc0 registers the client, users approve access on Arc0 Connect, and your agent reaches Longbridge through the same endpoint, policies and audit log as every other app.
Which Longbridge actions can my agent take?
161 in total: 1 read, 152 write and 8 destructive, such as “create watchlist group”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Longbridge?
Yes. Actions like “delete watchlist group” 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 Longbridge too?
Yes. The same Longbridge connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Longbridge into your agent.
Your users connect Longbridge once, under your brand. Your agent gets 161 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan