What is token-bleed AGENTS.md?

Repository instructions for Token Bleed, a local dashboard that reads Claude Code and Codex session logs and estimates token costs. It describes the project's commands, architecture, and data flow.

Use it when changing the TypeScript backend, log parsers, cost aggregation, dashboard server, or vanilla JavaScript frontend.

It helps coding agents understand where the logs come from, how they are processed, and how to build and run the dashboard correctly.

token-bleed AGENTS.md — Instructions file for Codex and OpenCode by MrRichSylvester: Repository instructions for Token Bleed, a local dashbo

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/mrrichsylvester/token-bleed/agents-md

Clone the repo

git clone --depth 1 https://github.com/MrRichSylvester/token-bleed

Made for: Codex, OpenCode.

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/mrrichsylvester/token-bleed/agents-md.svg)](https://agentmods.dev/instructions/mrrichsylvester/token-bleed/agents-md)

Your own site

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

Per session 816 This file is loaded in full into every session.

When invoked 816 The same file — it is already loaded in full.

Security scan B 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	$0.00816	$0.00816
Opus 5	$0.00408	$0.00408
Sonnet 5	$0.00163	$0.00163
Haiku 4.5	$0.00082	$0.00082

Measured 4d ago against content hash f4c0d4be1a66, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

token-bleed AGENTS.md scanned grade B 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 4d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

| `~/.claude/settings.json` | Claude Code settings (we read/write `cleanupPeriodDays`) |

AGENTS.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Guidance for AI coding agents (Codex, Gemini, etc.) working in this repository.

What this project is

Token Bleed is a local cost dashboard for Claude Code and Codex sessions. It reads .jsonl log files from ~/.claude/projects/ and ~/.codex/sessions/, computes token costs, and serves a web dashboard. No cloud, no telemetry, no accounts.

Commands

npm install          # install dependencies
npm run dev          # run with hot reload (tsx watch) — for development
npm run build        # compile TypeScript to dist/
npm run build:start  # run compiled output

Dev server runs at http://localhost:3000. The production CLI (npx token-bleed) runs at port 3847.

Architecture

The backend is TypeScript compiled to ESM (dist/). The frontend is vanilla JS served as static files from public/ — no build step, no framework.

Data flow

~/.claude/projects/**/*.jsonl  →  parser.ts  →  aggregator.ts  →  server.ts  →  API
~/.codex/sessions/**/*.jsonl   →  codexParser.ts  ↗

parser.ts reads Claude Code .jsonl files, extracts per-turn token usage from message.usage, resolves model names, and computes per-turn cost.
codexParser.ts does the same for Codex rollout logs.
aggregator.ts rolls parsed sessions into project, model, and daily summaries. All filtering by date/source happens here.
pricing.ts holds the model pricing table. Custom pricing set via Settings UI is merged in at runtime and persisted to ~/.burn-rate-settings.json.
providers.ts manages the model bridge feature: storing API keys (base64 in ~/.token-bleed/providers.json), spawning LiteLLM proxy processes, and tracking their PIDs.
server.ts is the Fastify app. All routes are in this one file. In-memory cache is invalidated on /api/refresh or every 5 minutes.

Frontend

public/app.js contains all UI logic as one file. Page views are rendered by functions (renderOverview, renderSessions, etc.) that fetch from the API and write into #content. State is a single top-level state object. Persistent UI state (hidden cards, view mode, column order) lives in localStorage.

Read the full file on GitHub · 60 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.

4d ago First seen · 60 lines · 816 tokens per session scan B f4c0d4be1a66

Subscribe to this mod's changes

token-bleed AGENTS.md is an instructions file published in the GitHub repository MrRichSylvester/token-bleed (11 stars, last pushed 1mo ago), licensed MIT. It adds 816 tokens to every session, about $0.0041 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

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-bleed AGENTS.md