What is Ataru AGENTS.md?

Repository-specific instructions for Codex, an AI coding assistant, covering the project's design rules, search model, common interface patterns, and development commands.

Use them when editing Ataru, a desktop app for finding useful context in AI coding-tool histories, particularly when working on its interface or search behavior.

They help the assistant make changes that match Ataru's existing structure and appearance instead of relying on assumptions.

Ataru AGENTS.md — Instructions file by lovstudio

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/lovstudio/ataru/agents-md

Clone the repo

git clone --depth 1 https://github.com/lovstudio/Ataru

Made for: Codex, OpenCode.

Per session 1,838 This file is loaded in full into every session.

When invoked 1,838 The same file — it is already loaded in full.

Security scan A 0 findings. 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	$0.01838	$0.01838
Opus 5	$0.00919	$0.00919
Sonnet 5	$0.00368	$0.00368
Haiku 4.5	$0.00184	$0.00184

Measured yesterday against content hash 198e05878a95, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Ataru AGENTS.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 yesterday.

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.

AGENTS.md · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

This file provides guidance to Codex (Codex.ai/code) when working with code in this repository.

Design System

This project uses Lovstudio Warm Academic Style (暖学术风格)

Reference complete design guide: file:///Users/mark/lovstudio/vault/design/design-guide.md

Quick Rules

禁止硬编码颜色：必须使用 semantic 类名（如 bg-primary、text-muted-foreground）
字体配对：标题用 font-serif，正文用默认 font-sans
圆角风格：使用 rounded-lg、rounded-xl、rounded-2xl
主色调：陶土色（按钮/高亮）+ 暖米色背景 + 炭灰文字
组件优先：优先使用 shadcn/ui 组件

Color Palette

Primary: #CC785C (陶土色 Terracotta)
Background: #F9F9F7 (暖米色 Warm Beige)
Foreground: #181818 (炭灰色 Charcoal)
Border: #E8E6DC

Common Patterns

主按钮: bg-primary text-primary-foreground hover:bg-primary/90
卡片: bg-card border border-border rounded-xl
标题: font-serif text-foreground

Project Overview

Ataru is a local-first AI conversation recall desktop app built with Tauri 2 + React 19 + TypeScript. It focuses on searching and recovering useful context from AI coding tool histories while retaining the legacy Lovcode CLI as a compatibility alias and preserving storage contracts.

Search Domain Model

Ataru 的公开搜索层级固定为 turn → run → session → project：

Turn：最小原子 transcript item，例如一条用户消息、AI 回复、工具调用或工具结果；Turn 不拥有标题。
Run：一次完整执行，从用户提出问题开始，到 AI 完成最终回应结束，包含多个 Turn。底层索引继续使用兼容字段 round_index / round_prompt，公开 API 使用 runIndex / runPrompt。
Session：同一来源的一段连续对话，包含多个 Run。
Project：一个工作目录及其历史 Session 集合。

UI 和搜索契约不得把 Turn 展示成“用户问题 + AI 完整回复”；那是 Run 的语义。Turn 结果应展示消息类型、正文、命中位置和所属 Run 上下文；Run 结果才可以使用用户首句作为标题。禁止为 Turn 或 Run 使用 Untitled turn、未命名回合 之类的标题占位。

Commands

# Frontend development (hot reload)
pnpm dev

# Type check + production build
pnpm build

# Run Tauri desktop app (auto-starts pnpm dev)
pnpm tauri dev

# Build distributable
pnpm tauri build

Architecture

Dual-layer architecture:

src/ - React frontend (Vite, port 1420)
src-tauri/ - Rust backend (Tauri 2)

Read the full file on GitHub · 107 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.

yesterday First seen · 107 lines · 1,838 tokens per session scan A 198e05878a95

Subscribe to this mod's changes

Ataru AGENTS.md is an instructions file published in the GitHub repository lovstudio/Ataru (354 stars, last pushed 8d ago), licensed Apache-2.0. It adds 1,838 tokens to every session, about $0.0092 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 2026-08-30.

Data and AI Frontend ReAct Research ai-tools claude-code desktop-app rust tailwindcss

Ataru AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

AGENTS.md

Design System

Quick Rules

Color Palette

Common Patterns

Project Overview

Search Domain Model

Commands

Architecture

What this file has done since we first saw it

Other instructions, from other repositories

orbit AGENTS.md

ship-studio AGENTS.md

codex-trace AGENTS.md

vibeshell AGENTS.md

vibeshell CLAUDE.md

claude-code-history-viewer CLAUDE.md