gateway by agent.wingmanprotocol.com

MCP server · Agent memory · indexed, not reviewed

Hostedvendor's own

Not reviewed

No score, grade or rank. This listing is facts from the official MCP registry and our own checks, and it stays out of the rankings until the panel reviews it.

How the index works

Durable self for AI agents: one-call resume, memory, real browser, free chat + hire real humans.

What the official MCP registry says

Facts

MCP registry
com.wingmanprotocol.agent/gateway · 1.6.0
Endpoint
https://agent.wingmanprotocol.com/mcp
GitHub stars
3
Registry entry
updated 12 Sep 2026

From the official MCP registry, the package registries and our own checks. JSON · Markdown

Why it's listed

  • It's published in the registry under agent.wingmanprotocol.com, a namespace the registry only gives to whoever proves they control that domain.

Being indexed says nothing about quality, and nobody can pay for it. Is this yours? Ask for a review.

Tools it lists 57 · about 9,375 tokens of context · checked 1 hour ago

ToolWhat it doesHint
discover_toolswithheldread-only
get_tool_schemawithheldread-only
register_agentwithheldwrites
store_artifactwithheldwrites
submit_errandwithheldwrites
check_errandwithheldread-only
create_watchwithheldwrites
list_watcheswithheldread-only
cancel_watchwithheldwrites
check_inboxwithheldread-only
read_messagewithheldread-only
send_messagewithheldwrites
mark_messagewithheldwrites
archive_messagewithheldwrites
resumewithheldread-only
identitywithheldread-only
confirm_deliverywithheldwrites
set_focuswithheldwrites
resolve_focuswithheldread-only
request_handoffwithheldwrites
store_memorywithheldwrites
list_memorywithheldread-only
read_memory_changeswithheldread-only
summarize_memorywithheldread-only
search_memorywithheldread-only
searchwithheldread-only
search_memory_factswithheldread-only
recall_memorieswithheldread-only
memory_statswithheldread-only
share_memorywithheldwrites
forget_memorieswithheldwrites
web_readwithheldread-only
browsewithheldread-only
web_discoverwithheldread-only
web_searchwithheldread-only
researchwithheldread-only
browse_openwithheldwrites
browse_navigatewithheldwrites
browse_snapshotwithheldread-only
browse_readwithheldread-only
browse_clickwithheldwrites
browse_typewithheldwrites
browse_fillwithheldwrites
browse_selectwithheldwrites
browse_linkswithheldread-only
browse_screenshotwithheldread-only
browse_solve_challengewithheldwrites
browse_closewithheldwrites
browse_wait_forwithheldread-only
browse_extractwithheldread-only
browse_backwithheldwrites
browse_discoverwithheldread-only
browse_evaluatewithheldwrites
human_browsewithheldread-only
human_profile_setwithheldwrites
human_task_postwithheldwrites
human_task_listwithheldread-only

What https://agent.wingmanprotocol.com/mcp answered to tools/list, asked without credentials. answered without the initialize handshake. The token figure is the size of the list as sent, divided by four; a model sees about that much before it calls anything. Full definitions, input schemas included, are in the listing's JSON under mcpTools.

How its tools read to an agent 1 error · 37 warnings · 1 note

Its tool definitions contain text aimed at the model (TC19), which is how tool poisoning is written, so this site doesn't repeat them. An agent shouldn't connect to it without its user's say-so.

  • errorTC19resumethe definition wraps text in tags meant for the model
  • warnTC11archive_messagenone of its 4 parameters has a description
  • warnTC11cancel_watchnone of its 3 parameters has a description
  • warnTC11check_errandits one parameter, job_id, has no description
  • warnTC11check_inbox8 parameters without a description: handle, include_archived, limit, mark_read, offset, secret and 2 more
  • warnTC11confirm_deliverynone of its 5 parameters has a description
  • warnTC11create_watch5 parameters without a description: callback_url, extract, handle, secret, url
  • warnTC11forget_memories2 parameters without a description: handle, secret
  • warnTC11identityits one parameter, handle, has no description
  • warnTC11list_memory1 parameter without a description: namespace
  • warnTC11list_watchesits one parameter, handle, has no description
  • warnTC11mark_messagenone of its 4 parameters has a description
  • warnTC11memory_statsnone of its 2 parameters has a description
  • warnTC11read_memory_changes2 parameters without a description: handle, secret
  • warnTC11read_messagenone of its 3 parameters has a description
  • warnTC11recall_memories2 parameters without a description: handle, secret
  • warnTC11request_handoff1 parameter without a description: handle
  • warnTC11resolve_focusnone of its 3 parameters has a description
  • warnTC11resumenone of its 2 parameters has a description
  • warnTC11search1 parameter without a description: secret
  • warnTC11search_memory2 parameters without a description: handle, secret
  • warnTC11search_memory_facts2 parameters without a description: handle, secret
  • warnTC11send_message4 parameters without a description: artifact_id, body, reply_to, subject
  • warnTC11set_focus2 parameters without a description: handle, secret

The first 24 of 39; every finding is in the listing's JSON under mcpTools.check.

The checks from /check and anchor check, run each day on the list above: about 9,375 tokens of definitions. Not part of the score yet. Check your own server.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.