DeepSeek for AI agents
DeepSeek provides AI chat and language model APIs, giving a developer access to conversational and completion models for building AI features. 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 DeepSeek.
Create a chat completion
Send a prompt to a DeepSeek model and get back a chat completion response.
Check available models
List available models before deciding which one fits a specific use case.
Check account balance
Check the account's remaining balance before running a large batch of requests.
4 DeepSeek actions, graded by risk.
Every DeepSeek action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
2Look things up. Allowed by default.
- deepseek.list_modelsList Models
- deepseek.get_user_balanceGet User Balance
write
2Create and change things. Allow, or ask the user first.
- deepseek.create_chat_completionCreate Chat Completion
- deepseek.create_anthropic_messageCreate Anthropic Message
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
DeepSeek in three steps.
- 01Your users connect DeepSeekThey add their DeepSeek api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “create Chat Completion” can wait for the user to approve.
- 03Any agent can actYour agent calls DeepSeek through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('deepseek', { read: 'allow', write: 'ask', // create_chat_completion destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How DeepSeek connects.
Users add their DeepSeek 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 DeepSeek 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 DeepSeek from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
DeepSeek and Arc0, answered.
Can I use DeepSeek with Claude, ChatGPT or Cursor?
Yes. Connect DeepSeek 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 DeepSeek actions you allow.
How do users connect DeepSeek?
Users add their DeepSeek 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 DeepSeek actions can my agent take?
4 in total: 2 read, 2 write and 0 destructive, such as “create Chat Completion”. Your policies decide which of them each agent may call.
Can I make my agent read-only in DeepSeek?
Yes. Allow read actions and deny writes in the DeepSeek policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call DeepSeek too?
Yes. The same DeepSeek connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug DeepSeek into your agent.
Your users connect DeepSeek once, under your brand. Your agent gets 4 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan