64 tokens across 1 tools — lean (< 1K). Measured 2026-08-16 under methodology v1.0.
| server (self-reported) | markitdown v1.8.1 |
| status | measured |
| tokenizer | tiktoken / o200k_base |
| launch command | uvx markitdown-mcp |
| isolation | docker · ghcr.io/astral-sh/uv:python3.12-bookworm-slim · network bridge · network enabled for package fetch; clean FS, no host credentials |
| env vars supplied | none |
| canonical SHA-256 | d708fc0ae4f3fb818d01cc3d4da0b74c2e5591a7982735d6200b3d56ab74d66b |
| category | vendor-official |
| source | https://github.com/microsoft/markitdown |
| tool | tokens | share | description | schema |
|---|---|---|---|---|
| convert_to_markdown | 62 | 96.9% | 18 | 31 |
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/markitdown/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.