Borrowing it
Nothing to install: this file belongs to lkh081231/screenshot-feedback-hook-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lkh081231/screenshot-feedback-hook-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/lkh081231/screenshot-feedback-hook-mcpWrote 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/lkh081231/screenshot-feedback-hook-mcp/claude-md)<a href="https://agentmods.dev/instructions/lkh081231/screenshot-feedback-hook-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/lkh081231/screenshot-feedback-hook-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/lkh081231/screenshot-feedback-hook-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/lkh081231/screenshot-feedback-hook-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.02933 | $0.02933 |
| Opus 5 | $0.01466 | $0.01466 |
| Sonnet 5 | $0.00587 | $0.00587 |
| Haiku 4.5 | $0.00293 | $0.00293 |
Grade A, and why
screenshot-feedback-hook-mcp 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 10d 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.
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 — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
本文件给在此仓库工作的 Claude Code 提供项目上下文与约定。
项目是什么
screenshot-feedback-hook-mcp —— 一个开源、跨平台（Windows / Linux / macOS）、低部署门槛的「截图反馈」工具，让 coding agent 在做前端设计、画工程图（EasyEDA/CAD）等任务时能看到自己产出的真实画面并据此自我纠正。 （曾用名 agent-eye，因 PyPI 相似度规则弃用；现在包名/命令名/分发名/仓库名统一为 screenshot-feedback-hook-mcp。）
核心技术现实（设计前提，勿推翻）：
- Claude Code 的 hook 只能回传文本，无法直接把图片塞进上下文（feature request anthropics/claude-code#16592 未实现）。
- MCP 工具能回传原生图片块（FastMCP 的
Image(data=..., format=...)）。 - DeepSeek Harness 没有这条限制：它有持久图片附件服务
ctx.attachments，原生插件可以在拦截点上把截图直接变成图片块推进上下文，agent 无需主动做任何事。 - 故采用三层架构：MCP 负责「agent 主动调用、直接拿图」；Claude Code hook 负责「操作后自动截图、回传路径让 agent 用 Read 读」；dsh 原生插件负责「手动 + 两个自动时机，直接推图片块」。
已锁定的设计决策
- 回传机制 = MCP server + Claude Code hook + DeepSeek Harness 原生插件 三层
- 技术栈 = Python ≥ 3.10 + uv/uvx
- 截图库 = mss（纯 ctypes 调原生 API，Linux 无需 imagemagick），压缩用 Pillow，MCP 用 mcp(FastMCP)
- MVP 截图范围 = 全屏 / 指定显示器；区域(坐标)、按窗口标题、URL/无头浏览器 = roadmap
- dsh 插件 = TypeScript（Cordis 插件），npm 名
dsh-screenshot-feedback-hook-mcp，钉死 dsh v0.1.0-rc.8；截图与压缩仍全部委托 Python CLI（capture --json），TS 侧只做 dsh 接线
目录结构（目标）
screenshot_feedback_hook_mcp/
├── core/capture.py # 截图核心：mss 抓帧 → Pillow bytes
├── core/optimize.py # 字节预算导向：降采样(~1568px)+迭代降 JPEG 质量至 ≤~80KB
├── core/platform_check.py # macOS 权限/Wayland 检测提示
├── server.py # MCP server（FastMCP，take_screenshot/list_monitors）
├── cli.py # CLI + 统一入口 entry()：无参数=MCP server，带子命令=CLI
└── __init__.py
examples/ # hook 配置样例、MCP 配置样例
tests/ # optimize 单测 + hook schema 单测 + capture 冒烟测试
pyproject.toml # uv 管理，[project.scripts] 只暴露 screenshot-feedback-hook-mcp
dsh-plugin/ # DeepSeek Harness 原生插件（独立 npm 包，双面：Host + 浏览器）
├── src/config.ts # Schemastery Config，同时充当 settings 命名空间的 schema
├── src/capture.ts # spawn Python CLI(--json) → ctx.attachments.saveImage
├── src/route.ts # 图片能力闸门 + 换 provider 提示文案
├── src/content.ts # 规范值 ↔ 图片块/插件来源消息 的共享投影
├── src/tools.ts # take_screenshot / list_monitors
├── src/auto.ts # tools/post-execute 与 agent/turn-stopping
├── src/index.ts # Host 半侧：installSettingsSection + 注册工具/自动时机
├── src/client/ # 浏览器半侧：设置页那张卡片（React）
│ ├── card-form.ts # 自己的暂存表单 + revision 设栅（不能 import dsh 那份）
│ ├── controller.ts # settings scope ↔ 表单，CARD_FIELDS 决定卡片长什么样
│ └── ScreenshotCard.tsx
├── tsdown.config.ts # 复刻 dsh 未发布的 clientBundle 产物格式
└── cordis.patch.yml # 组合包 patch 层（package.json 的 dsh.bundle 指向它）
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.
- 10d ago First seen · 107 lines · 2,933 tokens per session scan A e5aa041592a4
screenshot-feedback-hook-mcp CLAUDE.md is an instructions file published in the GitHub repository lkh081231/screenshot-feedback-hook-mcp (3 stars, last pushed 10d ago), licensed MIT. It adds 2,933 tokens to every session, about $0.0147 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 2026-08-31.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.