972 tokens across 3 tools — lean (< 1K). Measured 2026-08-16 under methodology v1.0.
| server (self-reported) | Microsoft Learn MCP Server v1.0.0 |
| status | measured |
| tokenizer | tiktoken / o200k_base |
| launch command | npx -y mcp-remote https://learn.microsoft.com/api/mcp |
| isolation | docker · public.ecr.aws/docker/library/node:22-slim · network bridge · network enabled for package fetch; clean FS, no host credentials |
| env vars supplied | none |
| canonical SHA-256 | 9e3f4a65f0dc136a6ffa5ed00f95785ec561e9a78413888da160e44ceb139346 |
| category | vendor-official |
| source | https://github.com/MicrosoftDocs/mcp |
| tool | tokens | share | description | schema |
|---|---|---|---|---|
| microsoft_code_sample_search | 396 | 40.7% | 163 | 111 |
| microsoft_docs_search | 297 | 30.6% | 129 | 42 |
| microsoft_docs_fetch | 277 | 28.5% | 196 | 30 |
Each tool is tokenized on its own, so the parts do not sum exactly to the whole: the array adds its own brackets and commas, and the tokenizer merges tokens across object boundaries. The badge number is always the count of the whole array, never a sum of parts.
npx -y mcp-context-cost verify results/microsoft-learn/measurement.json
That re-tokenizes the published capture and checks the count and the hash. If it disagrees with the badge, the badge is wrong — open an issue and it gets corrected.