486 tokens across 2 tools — lean (< 1K). Measured 2026-08-16 under methodology v1.0.
| server (self-reported) | exa-search-server v3.4.0 |
| status | measured |
| tokenizer | tiktoken / o200k_base |
| launch command | npx -y exa-mcp-server |
| 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 | EXA_API_KEY |
| canonical SHA-256 | eea558e4c905976a1bf6e014b1a1a73cffa2b9baa707c11da9ec26db70e38e46 |
| category | vendor-official |
| source | https://github.com/exa-labs/exa-mcp-server |
| tool | tokens | share | description | schema |
|---|---|---|---|---|
| web_search_exa | 289 | 59.5% | 113 | 119 |
| web_fetch_exa | 195 | 40.1% | 56 | 87 |
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/exa/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.