Real Email for CAMEL Role-Playing Agents

Bridge CAMEL's multi-agent role-playing with real email delivery — using MultiMail's oversight modes to safely transition from simulation to production.

CAMEL is a communicative agents framework for studying multi-agent behaviors through role-playing. Its agents take on specific roles and collaborate on tasks through structured dialogue. When these role-playing scenarios need to interact with the real world via email, MultiMail provides the safe bridge.

MultiMail's oversight modes are particularly valuable for CAMEL because role-playing agents may simulate business communications that shouldn't immediately become real emails. The gated_send mode lets agents compose emails naturally during role-play while a human decides which ones to actually deliver.

Integration uses CAMEL's tool system. Define MultiMail API functions as tools available to ChatAgent instances, giving role-playing agents email capabilities with built-in safety controls.

Why MultiMail

Built for CAMEL developers

Safe Simulation-to-Production Bridge

CAMEL agents role-playing business scenarios may draft realistic emails. MultiMail's gated_send mode lets these drafts exist as reviewable items rather than delivered messages, safely bridging simulation and real communication.

Role-Specific Email Permissions

Different CAMEL roles can have different email capabilities. An executive agent might have send access while a research agent only reads. MultiMail's per-mailbox oversight modes enable this fine-grained control.

Society-Level Communication Patterns

CAMEL's society framework models group interactions. MultiMail extends this to external email, letting agent societies coordinate on outbound communications while maintaining human oversight.

Research-Grade Audit Logging

Studying multi-agent email behavior requires detailed logs. MultiMail's audit trail records every email action with context, providing research-grade data on how role-playing agents handle communications.

Try it

Try it with your agent

No code, no dashboard. Paste this to your AI agent — it connects MultiMail, creates an inbox, and builds the flow for you.

You are my CAMEL implementation agent. Build a minimal, safe MultiMail email-capable CAMEL agent without inventing APIs or endpoints. 1. Get MultiMail ready: read https://multimail.dev/llms.txt, follow its instructions to connect the MultiMail MCP server, create a free inbox, and set up a verified sending domain. Use only the MCP tools and setup guidance exposed there. 2. Wire MultiMail into CAMEL using CAMEL’s real MCP integration: use CAMEL’s MCPToolkit with the MultiMail MCP server configuration, connect it, call get_tools(), and pass those FunctionTool objects into the CAMEL ChatAgent tools parameter. 3. Give the CAMEL agent three email abilities through the discovered MultiMail MCP tools only: check the inbox, compose draft replies, and send email. In the agent’s system message, make it bridge CAMEL role-playing into real email delivery by treating simulated collaboration as planning, then using MultiMail only when an email action is explicitly needed. 4. Send one test email from the verified sender to a developer-controlled test address. The message should identify itself as a MultiMail CAMEL integration test and ask the recipient to reply so the agent can confirm inbox-check and draft-reply behavior. 5. Run the agent in MultiMail gated_send oversight mode for all sending. Before anything sends, require the developer to review the exact recipient, subject, body, and timing. Do not switch to monitored or autonomous until the developer explicitly approves that change.
Getting started

Step by step

1

Create a MultiMail Account and API Key

Sign up at multimail.dev, create a mailbox, and generate an API key from your dashboard. Your key will start with mm_live_.

2

Install Dependencies

Install CAMEL and the HTTP library for calling the MultiMail API.

3

Create Email Tool Functions

Define Python functions for send_email, check_inbox, and reply_email that call the MultiMail REST API. Wrap each with CAMEL's FunctionTool.

4

Assign Tools to Agents

Create ChatAgent instances with email tools and system messages that explain the oversight mode. Only assign email tools to roles that need them.

5

Approve Pending Emails

Review pending emails in the MultiMail dashboard. In gated_send mode, no email is delivered until a human approves it.

FAQ

Common questions

Can CAMEL agents send real emails during role-playing?
Yes, but with safety controls. When a CAMEL agent invokes the send_email tool, MultiMail's gated_send mode queues the email for human review. The human decides whether the role-played email should actually be delivered, preventing accidental sends during simulations.
How do I limit which roles can send email?
Only pass email tools to the ChatAgent instances that should have email access. Agents without the send_email tool in their tool list simply cannot invoke it. You can also use separate mailboxes with different oversight modes for different roles.
Can I use CAMEL societies to study agent email behavior?
Yes. CAMEL's role-playing framework combined with MultiMail's audit logging provides a research environment for studying how agents handle email. The audit trail records every email action with full context, while gated_send prevents any unintended real-world side effects.
What oversight mode should I use for CAMEL experiments?
For research and testing, use gated_all so every email action (reads and sends) is logged and gated. For production deployments where the role-playing session handles real customer email, gated_send is appropriate — reads are free but sends need approval.
Related integrations

Explore more

AutoGen

multi-agent · python

View guide →

MetaGPT

multi-agent · python

View guide →

ChatDev

multi-agent · python

View guide →
Get started

The only agent email with a verifiable sender

Email infrastructure built for AI agents. Verifiable identity, graduated oversight, and a hosted MCP server. Formally verified in Lean 4.

Read the Docs →See Pricing