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 skills add KhazixW2/Everything-Maa --skill maa-pipeline-optiongit clone --depth 1 https://github.com/KhazixW2/Everything-MaaWrote 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/skills/khazixw2/everything-maa/maa-pipeline-option)<a href="https://agentmods.dev/skills/khazixw2/everything-maa/maa-pipeline-option"><img src="https://agentmods.dev/badge/skills/khazixw2/everything-maa/maa-pipeline-option/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/skills/khazixw2/everything-maa/maa-pipeline-option"><img src="https://agentmods.dev/badge/skills/khazixw2/everything-maa/maa-pipeline-option.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00081 | $0.03019 |
| Opus 5 | $0.00041 | $0.01510 |
| Sonnet 5 | $0.00016 | $0.00604 |
| Haiku 4.5 | $0.00008 | $0.00302 |
Grade A, and why
maa-pipeline-option 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 12d 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 — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pipeline Option 工作流
如果选项属于一个尚未定义起始状态、安全边界和验收条件的端到端自动化目标，先交给 $maa-workflow-build 建立任务契约；已有契约时，再用本 skill 完成选项接线与局部验证。
项目初始化接力
新增选项前先查目标项目根目录的 basic_info.md。存在且包含第 0 节时，读取“0. Maa Skills 接力协议”和第 1/2/3/6 节，先确认 interface、resource/task entry、目标节点与 Python 外部调用，再按本 skill 查实际 option surface 和读取路径。文档只是缓存：pipeline_override、context.get_node_data() 和 Custom 参数路径必须在当前文件中闭环核实。文件缺失或没有第 0 节时按本 skill 直接发现 option surface；不得自动调用 $maa-project-init，只有用户明确要求初始化或刷新时才调用。相关源码比文档新时视为缓存可能过期并以源码为准，不自动刷新或覆盖已有非空文档。
Option Surface 发现规则
从主 interface.json / interface.jsonc 出发，而不是从目录约定猜测。import[] 相对主 Interface 目录解析，顶层只向 Interface bundle 贡献 task、option 和 preset；task 条目自身可以携带 group、option 等展示引用。resource[].path 同样相对主 Interface 目录解析，目标 Pipeline 文件从这些资源根读取。
M9A 是根目录 interface.json + 根目录 tasks/**/*.json import + resource/base 与渠道资源组合；它的 agent 也是数组形式并声明 agent/bootstrap.py。assets/interface.json 与 assets/resource/** 不是 M9A 的布局，但它们是 MaaPracticeBoilerplate 系项目的常见有效布局；仍必须由主 Interface 的声明推导，不能按目录约定猜测。
TL;DR：先识别 option surface
新增一个 UI 选项需要先识别本项目实际声明的 option surface。选项可以定义在主 Interface，也可以定义在主 Interface import 的任务文件；task 注册处可能在主文件，也可能来自 import。不要按文件路径或 assets/... 约定猜测。
常见联动点如下，按项目实际协议取用，缺关键点会导致 UI 看不到选项或运行时读不到值：
| # | 位置 | 内容 |
|---|---|---|
| 1 | option 定义处 | 主 Interface 的 option 字典，或主 Interface import[] 指向文件里的 option 字典 |
| 2 | task 注册处 | 主 Interface 或 import 文件里的 task option: [] / group / preset 引用 |
| 3 | 主 Interface 声明资源根下的 pipeline/**/*.json |
预定义目标节点（pipeline_override 不会创建节点） |
| 4 | Python 代码（仅 Python 需要读取或执行 Custom 时必需） | context.get_node_data()、argv.custom_action_param、argv.custom_recognition_param 与 option 路径保持一致 |
⚠️ pipeline_override 只做属性合并，不会凭空创建节点。 少了第 3 步，
context.get_node_data()会返回None，运行时静默失败。
完整协议参考（嵌套 option、global_option、controller/resource 限制、占位符注入）：references/protocol.md
历史校正
- 不要把 pure override 当成唯一最佳解：只改已有节点字段时 pure override 最小；但涉及运行时事件库、计数、动态目标、识别后处理、失败策略、跨节点状态时，CustomAction/CustomRecognition 更合适。
- 不要把 Flag 节点当成唯一配置入口：M9A 的 v5 object-form 里，参数经常通过
action.param.custom_action_param、custom_action_param_code、recognition.param.custom_recognition_param进入 Custom；这和context.get_node_data("Flag")是不同通道。 - 字段路径必须闭环：UI 写哪条路径，Python 就读哪条路径；pure override 则 Python 不读，直接观察运行时行为。
enabled/enable不是审美选择：MaaFramework 原生启停用enabled；历史项目若已有enablehelper，可沿用并兼容，否则优先enabled。
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 185 lines · 81 tokens per session scan A bc298dfafe69
maa-pipeline-option is a skill published in the GitHub repository KhazixW2/Everything-Maa (12 stars, last pushed 4d ago), licensed MIT. It adds 81 tokens to every session and 3,019 once invoked, about $0.0004 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.
Other skills, from other repositories
develop-web-game
Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with rendergametotext.
isuzu-unity-cli
Control Unity Editor from the CLI with the isuzu-unity-cli command. Execute C# code, browse scene hierarchy, inspect/modify GameObjects, capture screenshots, read console logs, check compile status, control play mode, and execute menu items. Use when: user wants to interact with Unity Editor programmatically, run C#…
achievements
Steam-style achievement system with XP, levels, streaks, and skill trees. Gamifies the development workflow. 25 achievements across 5 categories.
battlenet-automation
Automate Battlenet tasks via Rube MCP (Composio). Always search tools first for current schemas.
clawd-animation-lite
A lightweight generator for short, 1–3 second pixel animations featuring Clawd, the Claude Code mascot, shown as a small coral-orange crab. It creates a complete HTML file that can be opened in a browser.
unreal-asset-operation
Query and manage UE assets via UCP. Use when the user asks about asset search, dependencies, referencers, asset CRUD, asset management, getting selected/current assets, opening/closing asset editors, or any editor asset operation in Unreal Engine.