ai-tab-optimizer CLAUDE.md

ai-tab-optimizer CLAUDE.md is an instructions file for coding agents from eiler2005/ai-tab-optimizer. It costs 1,027 tokens per session, scanned A, original, MIT.

Project-specific instructions for developing AI Tab Optimizer, including its browser extension and local AI server. They describe coding rules, commands, and the required development workflow.

In plain words
What is it for?
For installing, type-checking, building, running, and loading the extension, as well as starting and checking its local FastAPI server.
Why use it?
They give an AI coding agent the project's expectations, reducing inconsistent changes and missed checks.

Instructions file

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/eiler2005/ai-tab-optimizer/claude-md
Clone the repo
git clone --depth 1 https://github.com/eiler2005/ai-tab-optimizer

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.

agentmods badge for ai-tab-optimizer CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/eiler2005/ai-tab-optimizer/claude-md.svg)](https://agentmods.dev/instructions/eiler2005/ai-tab-optimizer/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/eiler2005/ai-tab-optimizer/claude-md"><img src="https://agentmods.dev/badge/instructions/eiler2005/ai-tab-optimizer/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,027 This file is loaded in full into every session.
When invoked 1,027 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.

ModelPer sessionOnce invoked
Fable 5 $0.01027 $0.01027
Opus 5 $0.00513 $0.00513
Sonnet 5 $0.00205 $0.00205
Haiku 4.5 $0.00103 $0.00103

Measured 4d ago against content hash fa95eb970bef, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ai-tab-optimizer 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 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.

Makes network callslowCapability

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

pnpm health # curl health check
CLAUDE.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Tab Optimizer — Claude Code Guide

Full product spec: @PROJECT.md | Dev setup: @SETUP.md | Feature scope: @MVP_FEATURES.md

Core Rules

  • No any — use unknown with type narrowing.
  • Typecheck after every changecd extension && pnpm typecheck. Stop on errors.
  • No speculative abstractions — don't create helpers unless there are 3+ call sites.
  • No comments unless logic is non-obvious. Never restate what code already says.
  • Follow existing patterns — Zustand actions, discriminated union messages, Tailwind classes, @shared/* imports.
  • Update docs when feature behavior changes.

Commands

# Extension
cd extension && pnpm install    # install deps
pnpm dev                        # watch mode → dist/
pnpm build                      # typecheck + vite build
pnpm typecheck                  # TS check only

# AI Server
pnpm server                     # FastAPI on localhost:8765
pnpm health                     # curl health check

# Load in Chrome
# chrome://extensions → Developer mode → Load unpacked → extension/dist/
# After service worker changes: click Reload on extension card

Workflow

Every non-trivial feature follows three phases — do not skip:

1. Research/research <topic> — reads code, writes docs/research/<slug>.md

2. Plan/plan <feature> — creates docs/plans/<slug>.md with task checklist. Do NOT code until approved.

3. Implement/implement <path-to-plan> — executes tasks, typechecks, marks done.

For small changes (typo, rename, single-line fix): skip planning, implement directly.

Skills (.claude/skills/)

Skill Trigger Purpose
/research Manual Deep-research a topic, output to docs/research/
/plan Manual Create implementation plan in docs/plans/
/implement Manual Execute an approved plan task by task

Subagents (.claude/agents/)

Agent Model Use case
code-reviewer Sonnet Review code for patterns, edge cases, project conventions
security-reviewer Opus Scan for injection, secrets, XSS, OWASP top 10
test-writer Sonnet Generate Vitest / pytest tests for given files

Read the full file on GitHub · 101 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.

  1. 4d ago First seen · 101 lines · 1,027 tokens per session scan A fa95eb970bef

Subscribe to this mod's changes

ai-tab-optimizer CLAUDE.md is an instructions file published in the GitHub repository eiler2005/ai-tab-optimizer (3 stars, last pushed 3mo ago), licensed MIT. It adds 1,027 tokens to every session, about $0.0051 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 2026-08-31.