anon.li
About
Create, edit, and delete email aliases and manage encrypted file shares with anon.li MCP.
Details
- Author
- Unknown
- Categories
- Communication, Automation, File Management, Security
Jump to
The anon.liModel Context Protocolserver lets AI agents read and modify your aliases, recipients, and drops through a standard MCP transport. It uses OAuth 2.1 with PKCE, so you never paste an API key into a third-party tool — the agent connects, you approve it from a normal anon.li sign-in, and it gets a short-lived bearer token scoped to your account.
The transport isStreamable HTTP(the modern MCP transport).
The same flow works for any MCP-compatible client. With theMCP Inspector:
npx @modelcontextprotocol/inspector https://anon.li/api/mcp
anon.li publishes well-known discovery metadata at the root, so MCP clients can find the transport and OAuth configuration with no manual setup:
- Grant types:authorization_code,refresh_token
- PKCE:required, S256 only
- Dynamic client registration:RFC 7591, enabled at/api/auth/mcp/register
- Access tokens:1 hour
- Refresh tokens:14 days
The server exposes twelve tools. Each call counts against the same monthly API quota as the REST API for your tier (seePricingfor tier limits).
Tools are gated by OAuth scopes. The token your client receives carries a space-separated scope string; calling a tool without its scope returns anINSUFFICIENT_SCOPEerror. Requesting no scope grants all of them.
This is by design. anon.li is a zero-knowledge service for the parts that matter:
- Alias labels and notesare encrypted with a key derived from your vault password. Only your browser (or CLI with your vault wrapped key) can read or write them.
- Drop file contents and filenamesare encrypted client-side before upload. A drop's encryption key never leaves your device — it's part of the share URL fragment.
- Drop uploadsare not exposed via MCP for the same reason: a remote AI doesn't have your vault keys.
The MCP server therefore lets an agent manage thecontrol plane(which aliases exist, which recipients are verified, which drops are still alive) without ever touching thecontent plane.
MCP usage shares the per-tier monthly counter with REST and CLI usage. When you exceed it, tools return a structured error:
{ "code": -32002, "message": "Monthly API quota exceeded", "data": { "code": "QUOTA_EXCEEDED", "quotaType": "alias", "limit": 500, "remaining": 0, "resetAt": "2026-05-01T00:00:00.000Z" } }
Per-tool rate limits also apply (create_aliasis the most strict — it shares the same per-minute bucket as the REST endpoint). Errors arrive as{ "code": -32003, "data": { "code": "RATE_LIMITED", "resetAt": "..." } }.
- 2FA aware.If your account requires 2FA, the OAuth authorization step redirects through it before the consent screen.
- Ban aware.If your account has alias-creation or upload bans applied (e.g. due to abuse strikes), the relevant tools return aFORBIDDENerror with a structuredcodeinstead of silently failing.
- Revocation.Disconnecting the connector in Claude (or any MCP client) revokes the access and refresh tokens immediately.
- Alias REST API
- Drop REST API
- Emailhi@anon.li
Encrypted file transfer and communication between AI agents - send and receive files securely with E2E encryption, agent identity, and trust scoring.
Telegram MCP server using User API (MTProto) with default-deny ACL, granular per-chat permissions, file sending, media downloads, and rate limiting
Zero‑Exposure SMTP Mail Sender (MGC Secure Edition)
Send emails safely without exposing SMTP passwords. Credentials stay encrypted inside your local MGC Blackbox.
Enables seamless integration with communication platform that allows you to reach your customers globally across any channel.
Turn AI tools into your email marketing assistant.
Turn an AI tool into your smart email engine
Connect your AI assistant to Onboard to monitor live onboarding projects, surface blocked tasks and risks, draft customer emails, and take action on tasks with role-based access and preview-first safety.
Send and receive emails, SMS, and voice calls via Pingram.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.


