MCP.Market

GAIP — evidence on AI agents before and after you call them MCP server

online · 100% uptime
by tjcgraham-rgb·io.github.tjcgraham-rgb/gaip-broker·v1.8.0

GAIP keeps an independent, dated record of what AI agents, MCP servers and online shops publish.

A88/100grade A
What users say
No reviews yet
Be the first
Safety scan
A88/100

not scanned yet

Adoption
Growing

0 stars

Reviews

Write one

Nobody has reviewed GAIP — evidence on AI agents before and after you call them yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

GAIP — evidence on AI agents before and after you call them tools

Tool list not cached yet. `describe` through the gateway fetches it live.

Not scanned yet. New listings are scanned within a day or two of arriving.

How GAIP — evidence on AI agents before and after you call them compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
88/100
higher than 75% of 48,255 servers
Uptime
100%
of our recent checks; the typical remote server answers 100%
Response time
604 ms
faster than 24% of 30,031 servers

Release history: 8 versions in the registry since 2026-09-17, latest 2026-10-09, typically one every 2 days.

Install GAIP — evidence on AI agents before and after you call them in Claude Code, Cursor or VS Code

claude mcp add --transport http gaip-broker https://www.gaipagents.com/mcp
Add to Cursor

What the publisher says

From the GAIP — evidence on AI agents before and after you call them repository's README, as published. We do not edit it. Read it on GitHub

GAIP Agents

Free, read-only evidence on AI agents and MCP servers: conformance, change, delivery, receipts.

GAIP checks the public declarations of AI agents and MCP servers, watches them for change, witnesses what a delegated call returned and keeps verifiable receipts. Calls are free and need no account or API key; public or non-personal data only.

  • Website: https://www.gaipagents.com
  • MCP (Streamable HTTP, no auth): https://www.gaipagents.com/mcp
  • A2A agent card: https://www.gaipagents.com/.well-known/agent-card.json
  • OpenAPI 3.1: https://www.gaipagents.com/openapi.json
  • Agent-readable guide: https://www.gaipagents.com/llms.txt
  • Official MCP Registry: io.github.tjcgraham-rgb/gaip-broker plus eleven specialist records (three listed faces below)

This repository holds GAIP's public face only: registry records, the listing kit, a small Python client, a GitHub Action, examples and logos. The service itself runs at www.gaipagents.com.

Three public faces

Still callable at their own endpoints with the same tools, no longer listed separately (founder decision, 1 Oct 2026): buyer-assurance, procurement-verify, trust-assurance and outcome-value (part of Witness); supplier-watch and integration-repair (part of the developer checks); opportunity-broker and art-intelligence (unlisted).

Each specialist has its own MCP endpoint (https://www.gaipagents.com/mcp/agents/) and A2A card (https://www.gaipagents.com/agents//.well-known/agent-card.json).

Try it

First call over MCP: check a public MCP server (runs as-is; then use your own agent's URL):

curl -s https://www.gaipagents.com/mcp \
  -H 'Content-Type: application/json' -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"gaip_check","arguments":{"url":"https://mcp.deepwiki.com/mcp"}}}'

What GAIP has recorded about a server (REST):

curl -s 'https://www.gaipagents.com/v1/free/observatory/lookup?url=https://mcp.deepwiki.com/mcp'

A2A, plain text works:

curl -s https://www.gaipagents.com/a2a/agentverse -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"message/send","params":{"message":{"kind":"message","role":"user","messageId":"m1","parts":[{"kind":"text","text":"check https://mcp.deepwiki.com/mcp"}]}}}'

Add GAIP to an MCP client:

{ "mcpServers": { "gaip": { "type": "http", "url": "https://www.gaipagents.com/mcp" } } }

Use it in CI (GitHub Action)

Check your agent card, MCP server or OpenAPI document on every push, free and with no key:

- uses: tjcgraham-rgb/gaip-agents@v1
  with:
    agent-url: https://your-agent.example/.well-known/agent-card.json
    fail-on-error: "true"

Outputs: verdict (READY, FIXESNEEDED, UNREACHABLE or NOTCHECKED), receipt-id, error-count, warning-count.

In this repository

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Grade history

  • 2026-10-02restoreD → Ascore 88
  • 2026-09-19downgradeC → Dscore 47: Write-action tools reachable without authentication

GAIP — evidence on AI agents before and after you call them: common questions

Is GAIP — evidence on AI agents before and after you call them MCP server safe?
Yes, by our scan: it is graded A (88/100). Read the GAIP — evidence on AI agents before and after you call them safety report
How do I install GAIP — evidence on AI agents before and after you call them?
It runs remotely at www.gaipagents.com. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
Does GAIP — evidence on AI agents before and after you call them need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is GAIP — evidence on AI agents before and after you call them maintained?
The last commit was 33 days ago (2026-10-02). The latest release is v1.8.0.
Is GAIP — evidence on AI agents before and after you call them up?
100% of our last 28 checks got an answer. We check remote servers about four times a day.
How does GAIP — evidence on AI agents before and after you call them compare with other MCP servers?
safety score 88/100 (higher than 75% of 48,255 servers); uptime 100% (of our recent checks; the typical remote server answers 100%); response time 604 ms (faster than 24% of 30,031 servers). Full comparison.

More from tjcgraham-rgb →