io.github.Digital-Threads/token-pilot
token-pilot
Save up to 80% tokens when AI reads code via AST-aware structural reading
alive probed 2026-10-14
25tools
8.2ktokens in Claude Opus 5.5
8.5ktokens in Claude Code
4.1 %of a 200k context window
4.8ko200k_base estimate
1.5 sto initialize
6 msto list tools
- Run it
- npx -y token-pilot@1.0.4 (stdio, probed in a container with the network off)
- Host
- npmjs.com
- Reports itself as
- token-pilot 1.0.4
- Capabilities
- tools
- Instructions
- 3 272 characters
- Claude count
- Claude Opus 5.5, measured 2026-10-14. Counted again when the tool definitions change. Method.
- Repository
- https://github.com/Digital-Threads/token-pilot
- Registry
- active, updated 2026-10-06 · registry entry
- Reachable
- 1 of the last 1 probe
Most expensive tools
| Tool | Tokens (o200k) | Share of this server |
|---|---|---|
| read_for_edit | 339 | 7 % |
| smart_read | 333 | 7 % |
| find_usages | 321 | 7 % |
| module_route | 295 | 6 % |
| test_summary | 289 | 6 % |
Badge and feed
[](https://mcpcost.ulehla.dev/s/io.github.Digital-Threads/token-pilot)
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.