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/text2future/flowix/claude-mdgit clone --depth 1 https://github.com/text2future/flowixWhat 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 | $0.09915 | $0.09915 |
| Opus 5 | $0.04958 | $0.04958 |
| Sonnet 5 | $0.01983 | $0.01983 |
| Haiku 4.5 | $0.00992 | $0.00992 |
Grade B, and why
flowix CLAUDE.md scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo xcode-select -r # 首次运行 How it starts
The opening of the file, as written. The whole thing — 630 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Flowix 是一款桌面笔记应用（Tauri 2 + Rust 后端，React 19 + TS + Tiptap 前端），内置 AI 代理（rllm v1.1，OpenAI / Anthropic / DeepSeek 全部走 openai_compatible provider）。
命令
export PATH="$HOME/.cargo/bin:/opt/homebrew/bin:/usr/local/bin:$PATH"
npm run tauri:dev # 推荐：独立 dev bundle ID (com.flowix.app.dev / "Flowix Dev")，可与生产 app 并存
npm run tauri:dev:win # Windows 开发启动：使用 app/flowix-desktop/tauri.windows.dev.conf.json
npm run tauri dev # ⚠️ 走默认 tauri.conf.json，与生产同 bundle ID (com.flowix.app)，已被生产占住时会立刻 exit 0
npm run dev # 仅前端 (localhost:1420)
npm run tauri:mobile:dev # 移动端桌面预览（等同 tauri:mobile:preview，不代表真机）
npm run tauri:android:dev # Android 真机/模拟器；首次先运行 npm run tauri:android:init
npm run tauri:ios:dev # iOS 真机/模拟器；首次先运行 npm run tauri:ios:init
npm run dev:mobile -- --port 1421 # 仅移动端前端预览 (localhost:1421)，不包含 Tauri 原生能力
npm run tauri build # 生产构建
npm run cli:build:dev # CLI debug build
npm run cli:build:prod # CLI production build for the current platform
npm run dsh:build:dev # Download the DSH CLI carrier and build the local dev runtime
npm run tauri:build:prod # Flowix production bundle + updater artifacts
pkill -f "node.*vite" 2>/dev/null # 端口冲突时
sudo xcode-select -r # 首次运行
移动原生工程由 Tauri CLI 生成，不手工维护 app/flowix-mobile/gen/android 或
app/flowix-mobile/gen/apple。Android 需要先安装 JDK、Android SDK/NDK 并设置
ANDROID_HOME、NDK_HOME；iOS 需要完整 Xcode（只有 Command Line Tools 不够）。
完成环境配置后分别运行：
npm run tauri:android:init
npm run tauri:android:dev
npm run tauri:ios:init
npm run tauri:ios:dev
Rust 测试（在 app/ 目录跑）：
cd app
cargo test -p flowix-core <module>::tests # 跑某 crate 某模块
cargo test -p flowix-core <module>::tests::test_xxx # 跑单个
cargo test --workspace --lib # 跑全部
Dev / Prod 并存打包
通过差异化 Tauri 配置，让 dev 版与已安装的生产版同时运行：
- dev：
npm run tauri:dev→app/flowix-desktop/tauri.conf.dev.json→ bundle IDcom.flowix.app.dev/Flowix Dev - 生产：
npm run tauri:build:prod→tauri.conf.json+ 平台覆盖层 + 签名覆盖层 → 平台专用tauri.*.production.local.json→ bundle IDcom.flowix.app/Flowix
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.
- 2d ago First seen · 630 lines · 9,915 tokens per session scan B f13a9642d8e2
flowix CLAUDE.md is an instructions file published in the GitHub repository text2future/flowix (381 stars, last pushed 2d ago), licensed MIT. It adds 9,915 tokens to every session, about $0.0496 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
rembric CLAUDE.md
Instructions for susomejias/rembric, covering claude.md, quick reference, claims need evidence, architecture and data access pattern.
unknowns CLAUDE.md
Instructions for hiendinhngoc/unknowns, covering claude.md, repo shape, skill conventions, title and adding or changing skills.
hermes-codex-operator AGENTS.md
Instructions for AlekseiUL/hermes-codex-operator, a project described as: A local-first operator that turns plain-language ideas into tracked, verified Codex work—planned by Hermes, executed by Codex, and explained back to you.
motion-superpowers CLAUDE.md
Claude Code instructions for ardha27/motion-superpowers, covering agent instructions: motion superpowers, 1. grounded research first, 2. deterministic frame timelines, 3. style rulebook enforcement and 4. visual test-driven development (tdd).
dexuse AGENTS.md
AGENTS.md instructions for yashau/dexuse, covering agents.md, product promise, stack, quality gates and versioning.
rembric AGENTS.md
Instructions for susomejias/rembric, a project described as: Persistent memory for AI coding agents — self-hosted, MCP-native, append-only. One Docker image, one SQLite file.