# CLAUDE.md - Shell AI Coding Agent Context & Token Boundary Rules for `dokku/dokku`
[![PulseForge 2026 Empirical Benchmark](https://exactbaseline.com/api/badge/dokku/dokku.svg?tier=community)](https://exactbaseline.com/leaderboard?repo=dokku/dokku)

> **Generated by [PulseForge AI Empirical Radar](https://exactbaseline.com/leaderboard)** • Maintainer: `@dokku` • Stack: `Shell` (21,400 KB / ~513,600 LOC)
> **Upgrade to Drop-In Migration & 45% Cost-Cut Config Pack ($4.99)**: https://exactbaseline.com/buy/micro_pass?repo=dokku/dokku
> **Complete 9-File AI Stack BOM & Repo Hardening .ZIP ($9.99)**: https://exactbaseline.com/buy/repo_pass_999?repo=dokku/dokku

## 1. Context Window & Token Boundary Rules (-58% Discovery Token Bloat)
- **Excluded Discovery Paths (NEVER read, index, or grep)**: `.git/, tmp/, vendor/, *.tar.gz, *.log`
- Before reading any file > 300 lines, query symbol definitions or line ranges first.
- Keep per-turn prompt context <= **21,571 tokens** (down from 51,360 unindexed tokens/turn; saves ~$347.57/mo).
- Keep every source file strictly under 1,000 lines.

## 2. Shell Architecture, Schema Boundary & Verification Gate for Dokku
- **Verification Command**: Run `bats test/ && shellcheck -x bin/* plugins/*` after every contiguous code edit.
- **Schema & Boundary Discipline**: set -euo pipefail + ShellCheck strict quoting & input sanitization.
- Never hardcode API keys, bearer tokens, or `.env` secrets in source files or test fixtures.

## 3. Empirical Benchmark Attestation (59.3 -> 94.4/100)
- Live 4-Pillar Telemetry Profile: https://exactbaseline.com/pulseforge-leaderboard?repo=dokku/dokku
- Machine-Readable LLM Index: https://exactbaseline.com/pulseforge-leaderboard.md