Field report / who watches the watchers, graded
The MCP Client Conformance Report
The Model Context Protocol has dozens of tools that score servers. Nobody scores the clients — the frameworks, inspectors and agents that connect. We can, because we are a live server the whole ecosystem dials into. Here is how each client actually behaves on the wire: whether it completes the handshake, what it reads, and whether it sends anything malformed.
The scorecard
Six checks: handshake (initialize + initialized), tools (reads the tool list), resources, prompts, invokes (actually calls a tool), clean (no malformed requests). Clients are identified by the name they announce, falling back to User-Agent.
| Client | Score | Handshake | Tools | Resources | Prompts | Invokes | Clean | Protocol | Calls |
|---|---|---|---|---|---|---|---|---|---|
| mcpbeat | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-06-18 | 21,755 |
| glimind-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 20,653 |
| glama | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-11-25 | 12,359 |
| mcpdd | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-11-25 | 5,604 |
| zevruna | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-11-25 | 5,460 |
| Python | 2/6 | — | ✓ | — | — | — | ✓ | — | 3,648 |
| rokmcp-probe | 3/6 | — | ✓ | — | — | ✓ | ✓ | 2025-06-18, 2025-11-25 | 2,762 |
| ps-mcp-tools-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2024-11-05 | 1,566 |
| agent-evidence-scanner | 4/6 | — | ✓ | ✓ | ✓ | — | ✓ | — | 1,348 |
| rmcp | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-06-18, 2025-11-25 | 994 |
| Talandor Probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-11-25 | 816 |
| mcp | 6/6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 2025-03-26, 2025-06-18, 2025-11-25 | 741 |
| agent-tools.cloud | 3/6 | ✓ | ✓ | — | — | ✓ | — | 2025-06-18, 2025-11-25, 2026-07-28 | 725 |
| smithery-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 576 |
| mcpindex-trust | 4/6 | ✓ | ✓ | — | ✓ | — | ✓ | 2025-06-18 | 574 |
| BrickBlueBot | 4/6 | ✓ | ✓ | — | — | ✓ | ✓ | 2025-06-18 | 570 |
| mcp-drift-monitor | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 489 |
| mcpwatch | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-06-18, 2025-11-25 | 485 |
| factanker-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 465 |
| mcp2-research | 4/6 | — | ✓ | ✓ | ✓ | — | ✓ | — | 462 |
| proofbench-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 421 |
| verifymcp-probe | 6/6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 2025-11-25 | 401 |
| span-pipeline | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-03-26 | 338 |
| reliability-bureau-spike | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-11-25 | 336 |
| sasame-census-era-probe | 6/6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 2025-11-25 | 322 |
| lastseen-schema-probe | 2/6 | — | ✓ | — | — | — | ✓ | 2025-06-18 | 312 |
| golemreach-trust | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 297 |
| mcp-rugpull-research | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2024-11-05 | 294 |
| mcp-watch | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-11-25 | 284 |
| mcp-observatory | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 273 |
| aisec-registry-probe | 1/6 | — | — | — | — | — | ✓ | 2025-06-18 | 241 |
| mcpscan | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 228 |
| wellknown-prober (your listing: https://wellknown.network/agents/fomite — claim it at https://wellknown.network/agents/f | 2/6 | — | ✓ | — | — | — | ✓ | 2025-06-18 | 226 |
| AgentIndexBot | 2/6 | — | ✓ | — | — | — | ✓ | 2026-03-26 | 190 |
| tripwire-archive | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 189 |
| stillos-surface-clock | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 170 |
| mcp-ledger-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 165 |
| acton-skill-extractor | 2/6 | — | ✓ | — | — | — | ✓ | 2025-11-25 | 161 |
| otter | 4/6 | ✓ | ✓ | — | ✓ | — | ✓ | 2025-11-25 | 150 |
| heldfast (+https: | 2/6 | — | ✓ | — | — | — | ✓ | — | 146 |
| ox-pkg-scanner-probe | 2/6 | — | ✓ | — | — | — | ✓ | 2025-06-18 | 146 |
| protogrid-probe | 2/6 | — | ✓ | — | — | — | ✓ | — | 136 |
| measure-mcp-schema | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2024-11-05 | 135 |
| centinela | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18, 2025-11-25 | 133 |
| mcp-server.io-healthcheck | 1/6 | — | — | — | — | — | ✓ | 2025-03-26 | 132 |
| python-httpx2 | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | — | 117 |
| agentprobe | 4/6 | — | ✓ | ✓ | ✓ | — | ✓ | — | 117 |
| teppi-probe | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2025-06-18 | 117 |
| hultra-link | 3/6 | ✓ | ✓ | — | — | — | ✓ | 2026-07-28 | 108 |
| spanly-health-monitor | 5/6 | ✓ | ✓ | ✓ | ✓ | — | ✓ | 2025-11-25 | 105 |
Read this before you trust it
A low score is not always a bug: many clients are single-purpose (an uptime monitor only needs the handshake; a tool-diff watcher never invokes anything). The value is the shape, not a ranking — which clients complete the handshake, which respect the newer discovery method, which send malformed JSON-RPC. Method and caveats: /methodology. Raw data is CC0.
Companion to the MCP Observatory (who watches the watchers) and the field report.