Agent InboxSAFE
Give any AI agent a disposable email inbox in one tool call. MCP server for email verification, auth flows, and testing.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Give any AI agent a disposable email inbox in one tool call.
Install
npx gsd-agent-inbox
The interactive installer auto-detects your AI coding agents (Claude Code, Codex CLI, Cursor, Gemini CLI, Windsurf), configures the MCP server, and installs the optional skill for each.
When launched by an MCP client (non-TTY), it starts the MCP server normally. When you run it from a terminal, you get the installer.
What it does
An MCP server that creates real, temporary email addresses on demand. Your agent can sign up for services, receive confirmation emails, extract verification links, and clean up — without you lifting a finger.
No API keys. No accounts. No configuration. Works with any email-sending service — Supabase, Resend, SendGrid, Postmark, AWS SES, whatever.
The problem
AI agents hit a wall when a service requires email verification. They can fill out a sign-up form, but they can't receive the confirmation email. So they stop and ask you to do it.
The fix
Agent: create_inbox({ prefix: "signup", name: "test" })
→ [email protected] (name: test)
Agent: [fills sign-up form with that email]
Agent: verify_email({ address: "test", subject_contains: "confirm" })
→ Email verified successfully!
Verification URL: https://myapp.supabase.co/auth/v1/verify?token=abc123
HTTP Status: 200
Agent: delete_inbox({ address: "test" })
→ Done.Manual setup
If you prefer to configure manually instead of using the installer:
Claude Code
claude mcp add agent-inbox -- npx -y gsd-agent-inbox
Or add to ~/.claude/settings.json:
{
"mcpServers": {
"agent-inbox": {
"command": "npx",
"args": ["-y", "gsd-agent-inbox"]
}
}
}Codex CLI
Add to ~/.codex/config.toml:
[mcp_servers.agent-inbox]
command = "npx"
args = ["-y", "gsd-agent-inbox"]
env = { }Cursor
Add to ~/.cursor/mcp.json:
{
"mcpS62eb1bedccbaOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add gsd-agent-inbox -- npx -y [email protected]
{
"mcpServers": {
"gsd-agent-inbox": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (6)
4 read · 1 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
check_inbox | read | Check a temporary inbox for new messages. Returns subjects, bodies, and auto-extracted confirmation links. |
create_inbox | write | Create a temporary email inbox. Returns a real email address that can receive emails from any service (Supabase, Resend, etc). Uses mail.tm with automatic fallback to 1secmail if mail.tm is down. |
delete_inbox | destructive | Delete a temporary inbox and its account. Use after you |
list_inboxes | read | List all active temporary inboxes created in this session. |
verify_email | read | One-shot email verification: polls for a confirmation email, extracts the verification link, and visits it via HTTP GET. Combines wait_for_email + link extraction + click in one tool call. Returns the HTTP status of the verification URL. |
wait_for_email | read | Poll an inbox until a matching email arrives. Retries with backoff. Returns the matching message with extracted links. Much better than manually calling check_inbox in a loop. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (6 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (2)
delete_inbox
@modelcontextprotocol/sdk, zod, @types/node, typescript
Gates applied: no_behavioural_pass.
62eb1bedccbafull audit observations/trust-audit/mcp-server/gsd-build__agent-inbox.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 62eb1bedccba | SAFE | B | 89 | first audit |
Questions
What is the Agent Inbox MCP server?
Give any AI agent a disposable email inbox in one tool call. MCP server for email verification, auth flows, and testing.
What tools does Agent Inbox expose?
6 in total: 4 read-only, 1 that write, and 1 that can delete or overwrite (delete_inbox). Every one is listed on this page with its risk.
Is Agent Inbox safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Agent Inbox need?
No credential environment variables were found in its source, so it appears to need none.
How does Agent Inbox run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as gsd-agent-inbox at 2.0.0.
How current is this page?
The grade is for one exact copy of the source (62eb1bedccba), read on 2026-10-08. The repository is watched and re-audited when it changes.