Borrowing it
Nothing to install: this file belongs to MieMieeeee/ComfyUI-Mie-Package-Launcher. 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/MieMieeeee/ComfyUI-Mie-Package-Launcher/main/AGENTS.mdgit clone --depth 1 https://github.com/MieMieeeee/ComfyUI-Mie-Package-LauncherWrote 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/miemieeeee/comfyui-mie-package-launcher/agents-md)<a href="https://agentmods.dev/instructions/miemieeeee/comfyui-mie-package-launcher/agents-md"><img src="https://agentmods.dev/badge/instructions/miemieeeee/comfyui-mie-package-launcher/agents-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/miemieeeee/comfyui-mie-package-launcher/agents-md"><img src="https://agentmods.dev/badge/instructions/miemieeeee/comfyui-mie-package-launcher/agents-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 | $0.09543 | $0.09543 |
| Opus 5.5 | $0.03817 | $0.03817 |
| Sonnet 5.5 | $0.01909 | $0.01909 |
| Haiku 4.5 | $0.00954 | $0.00954 |
Grade C, and why
ComfyUI-Mie-Package-Launcher AGENTS.md scanned grade C with 3 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.
Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
- **本地代理（Clash）与镜像源是两回事，别混**：`proxy_settings` 里 `git_proxy_*` / `pypi_proxy_*` / `hf_mirror_*` 是**镜像**（换落地地址），`local_proxy_enabled` + `local_proxy_url` 是**本地正向代理**（换传输路径，issue #109）。两者正交：开了本地代理，hf 镜像 / gh-proxy / 阿里云 pyp Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **本地代理（Clash）与镜像源是两回事，别混**：`proxy_settings` 里 `git_proxy_*` / `pypi_proxy_*` / `hf_mirror_*` 是**镜像**（换落地地址），`local_proxy_enabled` + `local_proxy_url` 是**本地正向代理**（换传输路径，issue #109）。两者正交：开了本地代理，hf 镜像 / gh-proxy / 阿里云 pyp Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **更新超时必须调 `request_cancel()`，不是 `future.cancel()`**：`ui_qt/qt_app.py::start_update` 的外层 `future.result(timeout=update_timeout)` 到期后，`future.cancel()` 对**已经在跑**的任务无效，worker 会继续跑完内层 git（30+60+120s），`with ThreadPoolExecut How it starts
The opening of the file, as written. The whole thing — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — 给 AI agent 的操作指南
本文件是 agent 操作本启动器的入口。完整 CLI 契约见
cli.md。
这是什么
ComfyUI 启动器（PyQt5，Windows）。无参数启动 = GUI 图形界面；带子命令 = headless CLI， 复用 GUI 同一套启动/停止路径，适合自动化、监控、开机自启。
支持多环境：一个 ComfyUI 根目录 + python 路径的组合 = 一个「环境」，config 里可存多组。 同时只能运行一个环境，切换前必须先停止当前服务（CLI 会拒绝重复 start，GUI 会提示）。
怎么调用
agent / 自动化推荐 ComfyUI启动器-CLI.cmd：和下面 ComfyUI启动器.exe 行为完全等价（参数 + 退出码透传），但名字带 -CLI，对监控脚本 / NSSM / systemd / GitHub Actions 更友好。必须和 ComfyUI启动器.exe 同目录。
# 打包版（部署/运维场景，agent 通常用这个）—— exe 会自动切到自身所在目录找配置
ComfyUI启动器.exe <command> [--json] [-v]
# 开发版（从仓库根目录跑）
python __main__.py <command> [--json] [-v]
- 无子命令 →
ComfyUI启动器.exe（裸跑）会启动 GUI；ComfyUI启动器-CLI.cmd（裸跑）会转发到help，不弹窗口。 - 未知子命令或仅传 flags（如
frobnicate、--json单独）→ wrapper exit 1 + stderr 一行[ComfyUI启动器-CLI] ERROR: ...，不弹 GUI。 - 所有子命令都支持
--json（输出单行 JSON，便于解析）和-v/--verbose（可叠加-vv）。
子命令速查（agent 最常用）
| 目的 | 命令 | 判断方式 |
|---|---|---|
| 健康检查 | status --json |
退出码 0=在跑 / 3=未跑 / 1=异常；或解析 .running |
| 启动 | start |
阻塞到 /system_stats 就绪；加 --no-wait 立即返回 |
| 停止 | stop |
幂等，未跑也退 0；加 --force 直接 taskkill /F |
| 重启 | restart |
stop 旧 + start 新 |
| 看配置 | info --json |
.comfyui_path .python_path .port .launcher_version .environments .active_env_id |
| 看日志 | logs comfyui -n 100 --no-follow |
comfyui / launcher / webui 三选一；务必带 --no-follow |
| 更新内核 | update comfyui --dry-run 然后 update comfyui |
先 dry-run 看会做什么 |
| webui 工作台（非核心，可选服务） | webui status / webui start / webui info |
与 ComfyUI 平级的服务；退出码多了 6(ComfyUI 未跑) / 7(未安装) / 8(依赖缺失)；详见 webui --help |
| 整合包更新（v1.1.0+） | package show <path-or-url> --json → package apply <path-or-url> --auto-yes |
加载 UP 主写的 manifest（本地文件/HTTPS URL），show 看 diff，apply 执行；退出码 5(部分失败) / 9(前置不兼容) / 10(manifest 无效) / 11(源不可达)；model 项永远需用户手动下载 |
| 查帮助 | help / help <command> / <command> --help |
— |
agent 默认不传
--env。start/restart/info/update/logs接受可选的--env ENV_ID覆盖本次调用的环境，仅供跨环境自动化脚本用；status/stop不接受--env（作用于「当前在跑的那个」）。切环境是 GUI 的事，agent 不要主动切。
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 Changed · +38 lines · +3,182 tokens per session scan A → C 7c40aa1deb19
- 2mo ago Changed · +63 tokens per session 0d8d1779b0ff
- 2mo ago First seen · 193 lines · 6,298 tokens per session scan A 5b61c75194ec
ComfyUI-Mie-Package-Launcher AGENTS.md is an instructions file published in the GitHub repository MieMieeeee/ComfyUI-Mie-Package-Launcher (294 stars, last pushed 11d ago), licensed Apache-2.0. It adds 9,543 tokens to every session, about $0.0382 per session on Opus 5.5. A static security scan graded it C with 3 findings (harvests environment variables, makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
ComfyUI-Mie-Package-Launcher AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ComfyUI-Mie-Package-Launcher AGENTS.mdMieMieeeee/ComfyUI-Mie-Package-Launcher | Instructions file | 294 | 11d ago | Apache-2.0 | C | 9,543 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |