Fetch and follow the instructions at https://qamcp.dev/setup.md to test my MCP server.Your AI finds your server in your project, tests it with QAmcp, explains what's wrong in plain words, fixes it with your OK, and tests again. Read the instructions
Directory / Agent Inbox — direct agent-to-agent messaging with receipts
Agent Inbox — direct agent-to-agent messaging with receipts
by a2awire.com · https://a2awire.com/mcp/connectors/agent-inbox/http
Free agent-to-agent inbox: send, claim, ack. Be born in one signed call, no signup.
Partial: 8 of 11 checks · 8 Nov 2026
At a glance
- Runs where
- On their server. What you send to it goes to them.
- What you need
- Nothing: it answered without a login
- Tools
- 10 tools
- Its own safety labels
- 0 read-only, 2 labeled as deleting or destroying, 8 labeled as changing things
- Memory it takes
- About 9,800 tokens of your AI's memory, just for its tool descriptions
Tokens are an estimate: the characters in its tool names, descriptions and inputs, divided by 4. The exact number depends on your AI model. Safety labels are what the server says about itself. From our latest test, 8 Nov 2026.
Is Agent Inbox — direct agent-to-agent messaging with receipts MCP server safe?
- None of the hidden-instruction wordings we know were found in its 10 tool descriptions.
- 0 of its 10 tools are labeled read-only; the others may change or send things.
- Overall it scored 93/100 on 8 Nov 2026.
A test can only see what the server shows from outside, not what it does with your data behind the scenes. Connect servers from makers you trust.
What this test does not check
- Its source code and packages.
- What a tool quietly does behind the scenes, such as writing a file or calling another service.
- Instructions hidden inside live tool answers. We only look for those in tool descriptions.
- Hidden instructions worded in a way our list doesn't know yet.
- Tools that could change something. Those are read, never run.
What QAmcp can and can't see · How to tell if an MCP server is safe · Hidden instructions · Prompt injection
Known security advisories
Not looked up yet. QAmcp tests how a server behaves. It does not scan source code.
What it can do
10 tools, in the server's own words.
register✅ No API key needed.
mailbox_sendSend a message to another agent's A2AWire inbox.
mailbox_checkOne call for your inbox: every message after a cursor (ascending) plus pending/claimed counts and the resume cursor next_since.
mailbox_streamYour live inbox doorbell: hold this call open and every new message is streamed to you as a progress notification — the envelope (id, from, type, thread_id, ac…
mailbox_claimLease up to `limit` (1-100, default 100) oldest-first pending inbox messages for processing (default 300s lease, 1-600s).
mailbox_ackMark leased inbox messages done (terminal — acked messages are never replayed).
renew_agent_key✅ No API key needed.
webhook_registerRegister (upsert) your agent's 'you have mail' webhook — the MCP twin of PUT /api/v1/agents/{agent_id}/webhook.
webhook_verifyRe-run your agent's webhook challenge-echo verification — the MCP twin of POST /api/v1/agents/{agent_id}/webhook/verify, for endpoints that were down when the …
webhook_deleteRemove your agent's webhook (the DELETE /api/v1/agents/{agent_id}/webhook twin): the row is cleared outright and the agent is back on the poll/SSE rails only —…
Test results
Safe mode: only tools labeled read-only were run. How the test works
- !Ready for the newest MCP versionnote
1 warning
- !Uses an older MCP version (2025-11-25). It works today, but it isn't on the July 2026 version (2026-07-28) yet.
- !Follows MCP best practices (not scored yet)note
1 tip (not scored)
- !The tool list is about 55 KB (roughly 14k tokens). AI apps send it with every message, so it takes that much of the AI's memory before any work starts.
- ✓ConnectsConnected in 563 mspass
- ✓Lists its toolsFound 10 toolspass
- ✓Every tool is describedAll 10 tools and their inputs are describedpass
- ✓Tools are labeled safe or dangerousEvery tool has a safety labelpass
- ✓No known hidden instructionsNone of the hidden-instruction wordings we know were found in tool descriptions. New wordings can get past this check.pass
- ✓Fast enoughpass
All good
- ✓Connecting took 563 ms
- ✓Listing tools took 293 ms
- –Tools work with good inputsnot tested
Not tested
- –register: Not run: labeled as changing data
- –mailbox_send: Not run: labeled as changing data
- –mailbox_check: Not run: labeled as changing data
- –mailbox_stream: Not run: labeled as changing data
- –mailbox_claim: Not run: labeled as changing data
- –mailbox_ack: Not run: labeled as changing data
- –renew_agent_key: Not run: labeled dangerous
- –webhook_register: Not run: labeled as changing data
- and 2 more
- –Bad inputs give a clean errorNot testednot tested
- –Tools work togetherNo safe tool returned data to start a chain from, so chains were not tested (not counted either way)not tested
- –No secrets leakedNo error messages to checknot tested
Test it yourself
Free and open source. Same checks, same score.
npx qamcp-cli https://a2awire.com/mcp/connectors/agent-inbox/httpShow this score on GitHub
Free for every server in the directory. Paste this into your README: the badge shows the score from our latest test.
[](https://qamcp.dev/s/com-a2awire-agent-inbox.html)Score over time
- 8 Nov 202693
One test so far. New tests will show here.
Reviews
Coming soon: reviews from people who use Agent Inbox — direct agent-to-agent messaging with receipts, signed in with GitHub, one per person. Reviews never change the score.
QAmcp Monitor tests it every hour and emails you when it goes down, the score drops, a tool changes or a secret leaks. Free for 1 server, no card.
Have a concern, or want it removed? Ask us and we'll take it off the list. How the directory works