CLAUDE CODE + ANIMA
Give Claude Code its own identity
The fastest way to give a Claude Code agent a real-world identity — its own email, phone number, voice, and credential vault — from your terminal. Add the MCP server, or paste one prompt and it signs itself up.
Connect
Connect Claude Code in one step
Pick a path — each gives Claude Code its own identity. The self-onboard prompt is the fastest; the MCP server and SDK wire it into your stack.
One command adds Anima's hosted MCP server to Claude Code.
claude mcp add --transport http anima https://mcp.useanima.sh/mcp \
--header "Authorization: Bearer ak_your_key"Why give it its own identity
Its own identity
Not your inbox or your number — your Claude Code agent gets its own. Every email, call, and login is attributable to the agent, not to you.
Compliant by default
TCPA, RND, and DNC checks run server-side before any call or text leaves the agent. You don't wire up compliance — it's on by default.
Use, never see
The agent authenticates with vaulted credentials it can use but never reads in plaintext, and clears its own 2FA challenges from SMS.
Working code
Prompt it, or drop in the SDK
Once connected, Claude Code acts as itself. Prompt it in plain language:
- “Sign up on Anima, then email me from your new inbox so I can verify you.”
- “Provision a US number and text me a confirmation from it.”
- “Check your inbox for replies, draft responses, and send the ones I approve.”
Provision an identity in code, then send from it.
import { Anima } from "@anima-labs/sdk";
const anima = new Anima(); // reads ANIMA_API_KEY
const agent = await anima.agents.create({
orgId: process.env.ANIMA_ORG_ID,
name: "My Agent",
slug: "my-agent",
provisionPhone: true,
});
console.log(agent.emailIdentities[0].email);
// my-agent@agents.useanima.shCapabilities
What your Claude Code agent gets
Email identity
A dedicated mailbox with DKIM, SPF, and DMARC — send and receive as itself.
Phone & SMS
A real US number to text from and to clear 2FA challenges.
Voice
Place and answer real-time calls on one high-quality pipeline.
Credential vault
Log in with secrets it can use but never sees in plaintext.
Compliance rails
TCPA, RND, and DNC gates run server-side on every call and text.
Unified audit
Every action carries a correlation ID back to the human who authorized it.
FAQ
Anima + Claude Code
- How do I connect Anima to Claude Code?
- Paste the self-onboard prompt and your agent reads useanima.sh/skill.md, signs itself up, and unlocks a full identity once you confirm the email — or point Claude Code at the hosted MCP server. No SMTP or provider setup.
- Do I need to run my own email or phone infrastructure?
- No. Anima provisions the inbox, handles deliverability (DKIM, SPF, DMARC), and supplies the phone number and voice pipeline. You just need an API key.
- Can the agent receive, not just send?
- Yes. Every identity can send and receive email and SMS, place and answer calls, and get webhooks in real time.
- Is there an MCP option?
- Yes — point Claude Code at https://mcp.useanima.sh/mcp with a Bearer key. The MCP server, CLI, and SDK all expose the same operations, so pick whichever fits your setup.
- How are credentials kept safe?
- Secrets live in an encrypted vault. The agent uses them to log in but never reads them in plaintext — reveal ≠ use — and every use is logged with a correlation ID.
- Is there a free tier?
- Yes. Start free with email included and no credit card. Phone, voice, and vault unlock on a paid plan.
Templates
Start from an example agent
Customer support agent
Answers inbound calls, verifies the caller over SMS, resolves on the line, and emails a recap — all from the agent's own identity.
View recipeOutbound SDR
Books meetings from the agent's own number — TCPA and DNC gates run server-side before every dial.
View recipeResearch assistant
Gathers and synthesizes — emailing sources for what the web won't give up, then delivering a cited report.
View recipe