com.completionkit/evals
CompletionKit
Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.
timeout probed 2026-10-15
fetch failed | Connect Timeout Error (attempted address: completionkit.com:443, timeout: 10000ms) | UND_ERR_CONNECT_TIMEOUT
- Endpoint
- https://completionkit.com/mcp (streamable-http)
- Host
- completionkit.com
- Repository
- https://github.com/homemade-software-inc/completion-kit
- Website
- https://completionkit.com
- Registry
- active, updated 2026-07-18 · registry entry
- Reachable
- 8 of the last 8 probes
Most expensive tools
| Tool | Tokens (o200k) | Share of this server |
|---|---|---|
| metrics_update | 800 | 13 % |
| metrics_create | 796 | 13 % |
| runs_create | 428 | 7 % |
| responses_list | 363 | 6 % |
| runs_update | 346 | 6 % |
History
We add a row when the result changes. A server with no change keeps its last row.
| Probed | Result | Init | Tools | Claude | o200k |
|---|---|---|---|---|---|
| 2026-10-15 09:33 | alive | 453 ms | 54 | 11k | 6.0k |
| 2026-10-14 09:10 | alive | 620 ms | 54 | 9.4k | 5.2k |
| 2026-10-05 13:04 | alive | 346 ms | 54 | 9.3k | 5.1k |
| 2026-10-05 00:12 | alive | 182 ms | 54 | 5.1k | |
| 2026-10-04 09:54 | alive | 856 ms | 54 | 5.1k | |
| 2026-10-04 07:48 | alive | 751 ms | 54 | 5.1k | |
| 2026-10-03 20:29 | alive | 908 ms | 54 | 5.1k | |
| 2026-10-03 17:47 | alive | 968 ms | 54 | 5.1k |
Badge and feed
[](https://mcpcost.ulehla.dev/s/com.completionkit/evals)
An entry for every probe that changed the result, the tool count or the token cost: Atom feed. Add it to a feed reader or a Slack RSS app to hear when the server stops answering or its tool set grows.