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.
npx agentmods add instructions/supremehui/-auto-accounting/claude-mdgit clone --depth 1 https://github.com/supremehui/-Auto-AccountingWrote 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.
[](https://agentmods.dev/instructions/supremehui/-auto-accounting/claude-md)<a href="https://agentmods.dev/instructions/supremehui/-auto-accounting/claude-md"><img src="https://agentmods.dev/badge/instructions/supremehui/-auto-accounting/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00598 | $0.00598 |
| Opus 5 | $0.00299 | $0.00299 |
| Sonnet 5 | $0.00120 | $0.00120 |
| Haiku 4.5 | $0.00060 | $0.00060 |
Grade A, and why
-Auto-Accounting CLAUDE.md scanned grade A with 0 findings 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 today.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository. Communicate in Chinese.
Project Overview
PinMe is an Android app (Kotlin, Jetpack Compose) that captures screenshots via a Quick Settings tile, uses vision LLMs to extract key information (pickup codes, train tickets, verification codes, etc.), and displays results through Flyme Live Notifications (Meizu-specific) or standard notifications plus a home screen widget.
Package: supreme
Min SDK: 33 | Target SDK: 36
Build Commands
Never use ./gradlew build in this project.
Architecture
Core Flow
- QuickCaptureTileService - Quick Settings tile triggers
CaptureActivity - CaptureActivity - Requests MediaProjection permission, captures screen as Bitmap
- ExtractWorkflow - Sends screenshot to vision LLM, parses JSON response
- UnifiedNotificationManager - Shows result via Flyme Live Notification (if available) or standard notification
- PinMeWidget - Glance AppWidget displays recent extractions
Key Components
| Package | Purpose |
|---|---|
capture/ |
Screen capture via MediaProjection API |
extract/ |
LLM workflow and JSON parsing |
vllm/ |
OpenAI-compatible API client (OkHttp) |
notification/ |
Notification handling with Flyme Live Notification support |
widget/ |
Glance AppWidget implementation |
db/ |
Room database for preferences and extraction history |
service/ |
LiveNotification service for Flyme integration |
LLM Configuration
Supports multiple providers configured in Constants.kt:
- 智谱 AI (ZHIPU) - Default, uses
glm-4v-flash - 硅基流动 (SiliconFlow) - Uses
Qwen/Qwen2.5-VL-72B-Instruct - Custom - User-specified OpenAI-compatible endpoint
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.
- today First seen · 67 lines · 598 tokens per session scan A 6101679a74d8
-Auto-Accounting CLAUDE.md is an instructions file published in the GitHub repository supremehui/-Auto-Accounting (7 stars, last pushed 7mo ago), licensed Apache-2.0. It adds 598 tokens to every session, about $0.0030 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other instructions, from other repositories
MarkdownView AGENTS.md
Instructions for keitaoouchi/MarkdownView, covering agents, technical components, relationships (data/event flow overview), extension points (overview) and platform / distribution.
droidconKeKotlin AGENTS.md
AGENTS.md instructions for droidconKE/droidconKeKotlin, covering agents.md, commands, module layout, stack and conventions.
skills copilot-instructions.md
Instructions for dhruvanbhalara/skills, covering project instructions & best practices, core tech stack, coding standards, architecture & data and ui & ux (design system).
compose-android-template CLAUDE.md
Claude Code instructions for ashtanko/compose-android-template, a project described as: 📱 A simple GitHub template to quickly set up an Android project using Kotlin and Jetpack Compose, getting you up and running in just seconds. Perfect for starting new apps with modern Android UI.
intentflow-ios AGENTS.md
Instructions for emrecanozturk/intentflow-ios, covering agent instructions, read first, architecture rules, commands and token budget.
SwiftfulStarterProject CLAUDE.md
Claude Code instructions for SwiftfulThinking/SwiftfulStarterProject, covering swiftfulstarterproject, architecture, build configurations, file creation and key conventions.