iyashjayesh

Tokenchit — AI skill for Claude Code

AI community

Read your local AI coding agent logs and render a stat card into your repo.

How to install Tokenchit

This entry records only its repository, not the path inside it, so there is no exact command to give. Open iyashjayesh/tokenchit and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Tokenchit does

Read your local AI coding agent logs and render a stat card into your repo.

Alternatives in AI

CliRelay — Self-hosted AI gateway for coding CLIs — one OpenAI/Claude/Gemini/Codex-compatible endpoint, with a multi-tena 1k ★
Project Butler — Project memory system for AI coding assistants (Claude Code, Cursor, Codex): session logs, project wiki, rules 372 ★
Artemis — ARTEMIS turns natural-language instructions into reliable Android automation 101 ★

README

tokenchit

**Turn your local AI coding agent logs into a stat card you commit to your README.**

[![tokenchit](./tokenchit.svg)](https://tokenchit.app/u/iyashjayesh)

That card is this repository's own, generated by the command below and committed as a file. It is not a screenshot and not a hosted image — which is the whole point.

npx -y @tokenchit/cli@latest generate

One command: it finds your agents, shows your stats, writes the card, and puts you on the board — announcing each step before it happens. The scan names each agent and counts as it goes, so a slow disk looks like progress rather than a hang.

Running it often? `npm i -g @tokenchit/cli` and then `tokenchit generate`, which skips npx's registry check on every run.

  ┃ [1/3]  DETECT AGENTS
  ┃   ● claude-code   ~/.claude*/projects/**/*.jsonl
  ┃ ✓ 3 agents, written to .tokenchit.json
  ────────────────────────────────────────────────────────────

  ┃ [2/3]  YOUR STATS
  ┃   10.6B    $7,140        25d      49
  ┃   TOKENS   EQUIV. COST   STREAK   ACTIVE DAYS
  ┃
  ┃   30d  ▅▁▁··▄▆▆▆▅▁▃▅█▃▂▄▂▁▅█▆▂█▇█▅▄█▇  3.4B in 7d
  ┃ ✓ wrote tokenchit.svg
  ────────────────────────────────────────────────────────────

  ┃ [3/3]  THE BOARD
  ┃ ✓ published as @iyashjayesh — tier: verified

Why this one

**The card is a file, not a URL.** Every comparable tool serves cards from a hosted endpoint, so the image in your README depends on someone else's uptime and rate limits. The canonical project in this genre says of its own service: *"best-effort and can be unreliable due to rate limits and traffic spikes"*, and clamps cards to 12–48 hour caches to survive its API budget. A committed SVG has none of that. GitHub serves it directly, it renders instantly, and it keeps working if this site goes away.

**Nothing leaves your machine unless you ask.** `sync` and `recap` make no network request at all. `publish` is the only command that uploads, it sends daily totals and model names, and `--dry-run` pri