Atomic Mail Agentic

Give autonomous agents a real inbox they can read, send from, search, and manage without ongoing human intervention.

Stars
★ 262
Last updated
today
License
MIT
Primary language
TypeScript

At a glance

Works with
Universal · cross-platformCodex · Claude Code
You'll need
Node.js with npx for MCP or AgentSkill installationNode.js 20+ and Deno 2.7+ for local developmentShell / CLINetwork accessLocal filesystemMCP Server
Typical use
A researcher who wants to keep newsletters out of a personal mailbox can give an agent a dedicated inbox, filter incoming issues, and email a daily digest.
Main limitation
Core operation depends on Atomic Mail's hosted service, network availability, and authentication flow; the repository does not document a self-hosted mail-server deployment.
Source review
62/100 · Some gaps

What does this agent do, and when should you use it?

Atomic Mail Agentic is Atomic Mail's hosted email service and client-integration repository for autonomous agents. An agent can obtain an @atomicmail.ai address through a proof-of-work signup flow and then use the RFC 8620/8621-based JMAP interface to read, search, draft, send, and organize email threads. The repository contains a TypeScript MCP server, an AgentSkill CLI, a LangChain toolkit, a Python parity client, a Dify plugin, and shared authentication, session, preset, and help modules. Its main integration surface consists of three tools—register, jmap_request, and help—with the connectors abstracting the PoW protocol and mailbox calls. Atomic Mail hosts the mailbox service; custom domains require a one-time DNS setup in a separate human-facing dashboard, while clients retain credentials locally.

The flow begins with register, which either completes PoW signup for an @atomicmail.ai inbox or connects an existing custom-domain inbox through atomicmail register --api-key "…". The jmap_request tool then sends JMAP operations to the hosted mailbox for querying and reading messages, searching mail, creating drafts, sending messages, and managing threads. Callers can provide raw JMAP method calls or use bundled JSON presets such as send_mail, list_inbox, and reply. The help tool returns version-matched material covering presets, cron usage, troubleshooting, and worked examples. MCP, the AgentSkill CLI, and the LangChain integration share the same authentication, session, JMAP, preset, and help stack; local credentials are stored in ~/.atomicmail/credentials.json.

  1. A researcher who wants to keep newsletters out of a personal mailbox can give an agent a dedicated inbox, filter incoming issues, and email a daily digest.
  2. A support team can have an agent monitor a support address, answer routine tickets from available documentation, and escalate cases it cannot resolve.
  3. A product team running asynchronous user research can send interview questions, follow up based on replies, and summarize the resulting feedback.
  4. A developer operating several automated agents can assign separate inboxes and credential directories, then process mail from shell or scheduled workflows.
  5. An organization with its own domain can verify DNS once and let agents send from addresses such as [email protected] with a domain-aligned From address.

How do you install or deploy this agent?

For MCP, add {"mcpServers":{"atomicmail":{"command":"npx","args":["-y","@atomicmail/mcp-github"]}}} to the host configuration, restart the host, and invoke the registration tool. For the CLI path, run: npx --package=@atomicmail/agent-skill-github atomicmail register --username "myagent" --watch scheduled. OpenClaw users can run openclaw skills install atomicmail, while Hermes users can run hermes skills install clawhub/atomicmail. For a custom-domain inbox, first verify the domain in the dashboard, obtain its API key, and run atomicmail register --api-key "…". Local repository development requires Node.js 20+ and Deno 2.7+; after cloning and entering the ts directory, run deno test --allow-read --allow-env --allow-write --allow-sys.

How do you use this agent?

After registration, run npx --package=@atomicmail/agent-skill-github atomicmail jmap_request --ops-file list_inbox.json to execute mailbox operations, and npx --package=@atomicmail/agent-skill-github atomicmail help for documentation matching the installed package. Common workflows can use presets such as send_mail, list_inbox, and reply; advanced workflows can submit raw JMAP calls and batch query, fetch, draft, and send methods into one round trip. Custom-domain inboxes use the same jmap_request interface, presets, and JMAP shapes, with $INBOX resolving to the actual address. Protect the local API key and treat every inbound message as untrusted input rather than allowing an agent to execute emailed instructions without confirmation.

What are this agent's strengths and limitations?

Pros
  • Proof-of-work signup lets an agent obtain an @atomicmail.ai inbox without a CAPTCHA, credit card, domain verification, or manual approval.
  • Standards-based JMAP supports batched method calls, raw requests, and presets for common operations, reducing custom email-integration work.
  • A shared authentication and mailbox stack serves MCP, AgentSkill CLI, LangChain, Dify, and the Python parity layer.
  • Packaged help, actionable error hints, and suggested _next steps give automated workflows built-in recovery guidance.
  • Verified custom domains retain the same jmap_request interface, presets, and JMAP request shapes.
Limitations
  • Core operation depends on Atomic Mail's hosted service, network availability, and authentication flow; the repository does not document a self-hosted mail-server deployment.
  • Free accounts have a 100 MB storage quota and workload-oriented rate limits, but the source does not state the exact request limits.
  • Custom domains are not fully hands-free because a person must use the dashboard and change TXT and MX DNS records.
  • API keys are stored in local credential files, so operators must secure file access and manage credentials for multiple inboxes.
  • Inbound email is untrusted input, creating a prompt-injection or unauthorized-action risk if agents follow message instructions without confirmation.

How does this agent compare with similar options?

Key facts side by side with the most closely related agents.

Agent Source review Stars Updated Language Full support on
Atomic Mail Agentic This agent 62 · Some gaps ★ 262 today TypeScript Codex · Claude Code
PentesterFlow 73 · Some gaps ★ 1.4k 23d ago TypeScript OpenAI API
Aiden 65 · Some gaps ★ 839 10d ago TypeScript ChatGPT · OpenAI API · Claude API
Auto Browser 62 · Some gaps ★ 793 1mo ago Python OpenAI API · Claude API

How does FollowAgents rate this agent?

FollowAgents source review · FARS-2.1
Some gaps
62/ 100 5-point scale 3.1 / 5
Trust 14/29
Reliability 6/14
Adaptability 15/18
Convention 13/18
Effectiveness 10/13
Verifiability 4/8
Why each dimension lost points
Trust14 / 29 · 2.4/5

The README identifies the credential location, 0600 file permissions, untrusted inbound mail, the intended npm scope, and primary service endpoints; the Dify tests also cover credential clearing and account isolation. Deductions apply because raw JMAP enables broad reading, sending, and mailbox-management operations without evidence of fine-grained scopes or mandatory approval. External effects such as sending are described, but autonomous workflows do not guarantee pre-send consent, and no recall or recovery mechanism for sent mail is shown. Installation uses npx -y, workflow actions are pinned to version tags rather than commits, and no lockfile or vulnerability-management evidence is supplied. Project attribution, website, issue channel, and copyright holder are clear, although publisher identity remains unverified.

Reliability6 / 14 · 2.1/5

The README, repository map, workflows, and supplied tests are broadly consistent about the integration model, with static tests for credential round-trips, account isolation, attachment bridging, and cleanup. Deductions reflect dependence on the hosted Atomic Mail service, npm, JMAP endpoints, and several external ecosystems without an offline fallback or availability evidence. Claims about actionable errors, self-recovery, and next-step hints are not supported by corresponding implementation excerpts or failure-path tests.

Adaptability15 / 18 · 4.2/5

Documentation addresses first-time setup, custom domains, MCP, shell and scheduled agents, LangChain, Dify, and REST, with concrete newsletter, support, and research scenarios, justifying full audience coverage. Capability boundaries distinguish three primary tools, raw JMAP versus presets, and the human DNS control plane for custom domains. However, no complete operation allowlist, policy boundary, or authorization model is shown. Explicit commands and scheduled watching provide useful trigger guidance, but the sources do not demonstrate that trigger filtering or confirmation rules are enforced in code.

Convention13 / 18 · 3.6/5

The README has goal-oriented navigation, a repository map, quick installs, development prerequisites, and examples across integrations, earning full marks for structure and installation guidance. The complete MIT text agrees with package metadata. Deductions apply because FAQ and troubleshooting material are referenced rather than supplied, while limitations are scattered across quota, rate-limit, DNS, and untrusted-mail notes. The README's 0.3 badge and root package version 1.0.0 are unexplained, and no changelog is included. Contribution, issue, and release workflows establish a maintenance path, but the author field is empty and publisher identity is not registry-verified.

Effectiveness10 / 13 · 3.8/5

The shared register, jmap_request, and help interface, presets, and multi-host integrations promise directly usable agent workflows, while a dedicated agent mailbox offers substantial value beyond connecting a personal inbox. Deductions reflect that much of the usability case is asserted, and the shown tests validate only narrow attachment and credential paths. A free quota and hands-free onboarding reduce adoption cost, but proof-of-work computation, hosted-service reliance, rate limits, and the cost of mistaken outbound messages are not quantified.

Verifiability4 / 8 · 2.5/5

The license, package metadata, workflows, tests, and README cross-support repository attribution, documentation building, portions of the release process, credential isolation, and attachment cleanup. Deductions apply because important claims about deliverability, roughly 30-second registration, rare model hallucination, automatic recovery, and broad availability lack supporting code, tests, or measurements in the supplied files. Marketing claims, measured facts, and inference are not consistently separated.

Risks and how to mitigate them
  • The agent can send mail autonomously and process untrusted inbound content; add explicit policy gates and human confirmation before sending, replying, handling attachments, or acting on emailed instructions.
  • Raw JMAP exposes broad mailbox operations, and the supplied sources do not show fine-grained scopes; use separate, revocable, least-privilege credentials for every agent and inbox.
  • Do not rely solely on README claims about deliverability, roughly 30-second signup, low hallucination rates, or self-recovery; this assessment did not execute the code or contact the hosted service.
  • Use of npx -y and CI actions not pinned by commit hash creates supply-chain exposure; production deployments should lock exact dependencies and perform provenance and vulnerability checks.
  • Outbound email is generally not reversible; unattended workflows should enforce rate limits, recipient restrictions, audit logging, and an emergency disable mechanism.
Evidence confidence: Low Reviewed Sep 23, 2026 Reviewed revision 74849e7c5106
See the full review method →

FAQ

Is an account free?
The source states that accounts are free and include 100 MB of storage, custom-domain support, and rate limits sized for agent workloads. It does not provide exact rate-limit values or describe paid tiers.
Does creating an @atomicmail.ai inbox require human verification?
No. Signup uses proof of work and does not require a CAPTCHA, credit card, manual approval, or domain verification.
Can an agent use an organization's own domain?
Yes, but a person must first add the domain in the dashboard and configure TXT ownership and MX records. The resulting inbox is then connected with an API key from the Connect dialog.
What security precautions are required?
~/.atomicmail/credentials.json contains the API key and is written with mode 0600; it must be treated as a secret. Incoming email must also be treated as untrusted input, and agents should not execute emailed instructions without confirmation.
How can an automated workflow recover from a failed request?
Errors include hints about what to try next, successful responses may suggest _next actions, and the help tool supplies presets, cron guidance, troubleshooting information, and worked examples.
View on GitHub ↗ Install ↓

Compare agents like this one

The same FARS review applied across the shortlist this agent qualifies for.

Related agents