What is token-dashboard CLAUDE.md?

Project instructions for a local dashboard that tracks Claude Code usage, costs, sessions, prompts, projects, skills, and related analytics. Claude Code is a coding assistant, and the dashboard reads its local JSONL transcript files and stores results in SQLite, a local database format.

Use it when working on the Token Dashboard, including its Python scanner and server, JSON APIs, live updates, or vanilla JavaScript interface.

It gives an agent the project context, architecture, supported platforms, and checks needed when editing the repository. This reduces the chance of changes conflicting with the existing dashboard design.

token-dashboard CLAUDE.md — Instructions file for coding agents by nateherkai: Project instructions for a local dashboard that tracks Claude

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/nateherkai/token-dashboard/claude-md

Clone the repo

git clone --depth 1 https://github.com/nateherkai/token-dashboard

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/nateherkai/token-dashboard/claude-md.svg)](https://agentmods.dev/instructions/nateherkai/token-dashboard/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/nateherkai/token-dashboard/claude-md"><img src="https://agentmods.dev/badge/instructions/nateherkai/token-dashboard/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 787 This file is loaded in full into every session.

When invoked 787 The same file — it is already loaded in full.

Security scan A 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00787	$0.00787
Opus 5	$0.00394	$0.00394
Sonnet 5	$0.00157	$0.00157
Haiku 4.5	$0.00079	$0.00079

Measured 6d ago against content hash b01faad9b455, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

token-dashboard CLAUDE.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl http://127.0.0.1:8080/api/overview # sanity-check an endpoint

CLAUDE.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Guidance for Claude Code when working in this repository.

Project overview

Token Dashboard — a local dashboard for tracking Claude Code token usage, costs, and session history. Reads the JSONL transcripts Claude Code writes to ~/.claude/projects/ and turns them into per-prompt cost analytics, tool/file heatmaps, subagent attribution, cache analytics, project comparisons, and a rule-based tips engine.

Inspired by phuryn/claude-usage but diverges in UI (vanilla JS + ECharts, dark theme, hash router, SSE refresh) and scope (expensive-prompt drill-down, skills view, tips engine, streaming-snapshot dedup). See docs/inspiration.md for the original's feature set and known limitations.

Status

Working codebase. 68 Python unit tests (python3 -m unittest discover tests). Seven UI tabs wired up (Overview, Prompts, Sessions, Projects, Skills, Tips, Settings). Runs on macOS, Windows, and Linux.

Architecture

cli.py → token_dashboard/scanner.py → ~/.claude/token-dashboard.db (SQLite)
token_dashboard/server.py exposes JSON APIs (/api/*) + SSE stream (/api/stream) + static frontend (web/)
web/ is vanilla JS, no build step — hash router + ECharts

Data source

Claude Code writes one JSONL file per session to ~/.claude/projects/<project-slug>/<session-id>.jsonl. Each line is a message record; usage fields live at message.usage and model identifier at message.model. The scanner is incremental — it tracks each file's mtime and byte offset in the files table and only reads new bytes on subsequent scans.

Conventions

Fully local. No telemetry, no remote calls for user data. Tests run offline.
Stdlib only. No pip install. If a new feature needs a third-party library, argue for it first — we're willing to pay ergonomics cost to keep install friction at zero.
SQLite parameter binding always. Any f-string in a SQL statement must interpolate only internal, caller-controlled values (column names, placeholder lists). User-reachable values go through ?.
Small files with clear responsibilities. If a file grows past ~400 lines or accretes three distinct concerns, split it.
Streaming-snapshot dedup. When adding scanner logic that joins the messages table, remember (session_id, message_id) is the dedup key, not uuid. See scanner._evict_prior_snapshots and the migration note in db._migrate_add_message_id.

Read the full file on GitHub · 48 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

6d ago First seen · 48 lines · 787 tokens per session scan A b01faad9b455

Subscribe to this mod's changes

token-dashboard CLAUDE.md is an instructions file published in the GitHub repository nateherkai/token-dashboard (685 stars, last pushed 4mo ago), licensed MIT. It adds 787 tokens to every session, about $0.0039 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Analytics and metrics

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

token-dashboard CLAUDE.md