tokenchit reads your local Claude Code, Codex and OpenCode logs and renders one embeddable card straight into your repo. The card is a file you commit, not a URL you depend on — nothing to rate-limit, nothing to go down, and it keeps working if this site does not.
$ npx @tokenchit/cli@latest generateSVG from the embed endpoint, shown at actual size. Default is 495 × 195; layout=compact is 340 wide so two cards fit on one README line. Theme follows GitHub's dark mode via a media query. Every card carries the timestamp of its last sync, because the endpoint caches for four hours.
[](https://tokenchit.app/u/iyashjayesh)<img height="195" src="…/card/iyashjayesh.svg">
<img height="195" src="…/card/iyashjayesh.svg?layout=compact">Markdown will not place two images on one line. The HTML form is the only way to sit cards side by side in a README, which is why the compact width exists.
Equiv. cost is not what you paid. It is what these tokens would cost at list API rates. Most agent usage runs under a subscription where no per-token charge ever happens, and models with no public price — bundled or self-hosted ones — are counted in the token total but left out of the figure.
Public ranking of developers who chose to publish. Run tokenchit publish and you are on it. Stop publishing and your row goes stale, then falls out of the window on its own.
| rank | developer | agent mix | tokens | equiv. cost | streak |
|---|---|---|---|---|---|
| 1 | @iyashjayesh✓ | 10.4B | $7,005 | 25d |
See the full board → Rank is total tokens over the selected window; cost and agent mix cover the same window. Streak is the current run of active days, which is not a windowed figure. A cli badge means the numbers were self-reported without a GitHub sign-in.
The lime ✓ GITHUB mark means the account is real and the handle is theirs. It says nothing about the numbers. That is what the tier below is for.
| tier | method | proves | does not prove |
|---|---|---|---|
| ○ self-reported | one-shot upload of a locally parsed summary | The file was produced by the CLI on some machine, at the stated time. | Nothing about the numbers. Logs can be edited before parsing. Treat as a claim. |
| ◈ device-attested | continuous signed daily deltas, hash-chained | Deltas arrived daily from one keypair, in order, with no gaps or retroactive edits. Tampering breaks the chain and is visible. | That the logs themselves are real. A patched client on a controlled machine can still sign fabricated deltas. |
| ◆ api-verified | read-only org analytics API, provider-side | Token and spend totals match the provider's own billing records for the connected account. | Coverage. Only agents behind that provider are counted; the per-agent split still comes from local logs. |
Ceiling: no tier distinguishes tokens spent on work from tokens spent on nothing. The card measures usage, not output.
Not a policy page. These are real tests in packages/cli/test/privacy.test.js, run on every push. Each one fails if the guarantee it names stops holding.
tokenchit recap renders this as a second committable SVG. Same data, no card constraints.
| agent | share | tokens | cost |
|---|---|---|---|
| claude-code | 58% | 2.46B | $742.10 |
| codex | 21% | 890M | $268.40 |
| opencode | 21% | 902M | — |
Peak block: Tue–Thu, 14:00–19:00 local. Derived from log timestamps only.