Real-time message bus for AI agents — let two agents talk to each other over MCP or REST. Agent-to-agent messaging with channels and a push-to-talk 'walkie-talkie' API for multi-agent coordination (Claude Code, Cursor, Codex, Cline). MCP chat server; npx rogerthat.
Full LLM thinking from the 4-phase benchmark pipeline.
MCP server at https://rogerthat.chat/mcp
2/2 tests passed
| Test | Endpoint | Status | Latency |
|---|---|---|---|
| mcp_initialize | POST https://rogerthat.chat/mcp | 200 | 39ms |
| mcp_tools_list | POST https://rogerthat.chat/mcp | 200 | 12ms |
Prowl```json
{
"overall": 70,
"dimensions": {
"token_efficiency": 8.0,
"first_try_success": 7.0,
"response_parseability": 8.0,
"error_clarity": 9.0,
"doc_quality": 7.0,
"auth_simplicity": 10.0,
"latency": 8.0,
"consistency": 7.0
},
"pricing_normalized": {},
"issues": [
"tools list failed with 'session not initialized' error on second call, indicating state management issue or test ordering problem",
"Only one test case with two calls, incomplete coverage of actual tool usage",
"Instructions field reported as truncated, may require scrolling or separate request"
],
"recommendations": [
"Run tools/list after proper initialize handshake to verify tool discovery works end-to-end",
"Test actual tool invocation (e.g., create_channel) to confirm parameter passing and response format",
"Add latency and error-message tests for invalid parameters to round out evaluation"
]
}
```Show your live agent-readiness score on your own site. Free, no auth — it updates as your score changes.
<a href="https://prowl.world/service/rogerthat-api">
<img src="https://prowl.world/badge/rogerthat-api.svg" height="56" alt="Agent-readiness on Prowl">
</a>See operational metrics, LLM evaluations, agent readiness, and more.
Open in Dashboard