Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@yf-decide-mcp Pick the right support team for 'my order never arrived, please refund' and rate urgency."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
一个通用的 MCP 服务，对接 System One 类型化决策端点（POST {BASE_URL}/v1/systemone）。
它只做一件事：让你能从任何 MCP 客户端使用任意 System One 提供方，包括那些
jev-decision-mcp 用不了的。
那个服务把 usage.output_tokens 当作必填字段，缺失就返回 INVALID_RESPONSE；
百炼的 decision-model-preview 只返回 input_tokens，所以在那边根本跑不起来。
本服务对 usage 做宽松归一化（缺的字段补 null），其余原样透传。
不内置任何分类体系或业务判据——问题由调用方每次请求时传入。
npx -y yf-decide-mcpRelated MCP server: laya-mcp
变量 | 必填 | 说明 |
| 是 | 端点根地址。不要带 |
| 是 | 默认模型名，可被单次调用覆盖。 |
| 否 | Bearer 凭据。不设时请求里完全不带这个头。 |
| 否 | 单次请求超时，默认 |
MCP 客户端配置：
{
"mcpServers": {
"systemone": {
"command": "npx",
"args": ["-y", "yf-decide-mcp"],
"env": {
"BASE_URL": "http://127.0.0.1:3425",
"API_KEY": "your-key",
"MODEL": "your-model"
}
}
}
}要同时对接两个提供方，就把这个服务注册两次、起不同名字、给不同的 BASE_URL / MODEL——
每个进程固定一个端点。
decide 工具只有一个工具，四种问题类型：
| 回答什么 | 返回字段 |
| 从 |
|
| 落在 2–10 个有序档位上的位置，0 起 |
|
| yes 的概率（0–1） |
|
| 感兴趣区域，归一化 0–1 |
|
{
"name": "decide",
"arguments": {
"state": { "message": "我的订单一直没到，帮我退款" },
"questions": {
"team": {
"type": "choice",
"instructions": "这条消息该由哪个团队处理？",
"criteria": {
"billing": "付款与退款",
"shipping": "配送",
"other": "其他"
}
},
"urgency": {
"type": "score",
"instructions": "这件事有多紧急？",
"criteria": ["可以等", "本周内", "今天", "立刻"]
}
}
}
}返回：
{
"model": "typesafe/jev-1.14-20261014",
"answers": {
"team": {
"type": "choice",
"choice": "billing",
"probabilities": { "billing": 0.87, "shipping": 0.12, "other": 0.01 },
"confidence": 0.8
},
"urgency": {
"type": "score",
"score": 1.64,
"legend": { "0": "可以等", "1": "本周内", "2": "今天", "3": "立刻" },
"probabilities": { "0": 0.02, "1": 0.4, "2": 0.5, "3": 0.08 },
"confidence": 0.48
}
},
"usage": { "input_tokens": 398, "output_tokens": 53 },
"provider": "TypeSafe",
"id": "gen-dec-..."
}把模型需要的事实全放进 state。它看不到你的对话，也读不到本地文件。
完整的判断写在 instructions 里。问题 ID 只是返回时的键名，模型不认它。
criteria 的描述要具体。写成「敏感 / 不敏感」这种，概率会糊在 0.5 附近。
需要时给一个「都不匹配 / 其他」选项。
做阈值判断时看概率，别看 choice。0.55 和 0.98 在业务上完全不是一回事。
score 是期望档位，即 $\sum_i i \cdot p_i$，不是强度大小。要结合 legend 读。
noul 接近 0.5 表示对 yes/no 不确定，不是「中等强度」。
失败一律返回 isError: true 加一段 JSON，绝不编造答案：
错误码 | 含义 |
| 端点连不上 |
| 超过 |
| 返回不是 JSON，或没有 |
| 上游返回非 2xx（错误信息会带出来） |
| 工具名不对 |
注意：部分提供方自带输入审核，可能在模型看到请求之前就用 4xx 拒掉。这会被归为
API_ERROR——请把它当作提供方的策略结果，而不是传输故障。
jev-decision-mcp 的差异jev-decision-mcp | systemone-mcp | |
| 必填 | 可选，缺了补 |
不返回该字段的端点（如百炼） |
| 正常可用 |
内置分类体系 | 无 | 无 |
运行时依赖 | MCP SDK + zod + TypeSafe SDK | 仅 MCP SDK |
问题类型 | choice / score / noul | choice / score / noul / box |
MIT
Ask one or more typed questions about a piece of context through a System One decision endpoint (POST /v1/systemone). This is a decision model, not a chat model: it does not generate prose, it answers the questions you define.
Question types:
choice: pick exactly one label from criteria (an object mapping label -> description).
score: position on 2-10 ordered levels given as an array, 0-based. Returns score, legend, probabilities and confidence.
noul: probability of yes, 0-1. criteria may optionally describe {true, false}. A value near 0.5 means uncertainty about yes/no, not medium intensity.
box: region of interest as {top,bottom,left,right} normalized 0-1.
Batch independent questions in one call; questions cannot see each other's answers. The model cannot see your conversation or local files — put every fact it needs into state, and write each full judgment into instructions. Question IDs are only response keys.
Inputs: state (required), questions (required), model.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 1 tool
There is only one tool, so there is no possibility of confusing it with another tool. Its purpose is clearly stated as a typed decision endpoint rather than a chat model.
The single tool name 'decide' is a clear, concise verb that matches the server's decision-making purpose. With only one tool, there are no inconsistent naming conventions to conflict.
A single tool is appropriate for wrapping one focused decision endpoint, and the rich question-type schema keeps it from being trivial. However, the surface is very thin, with no separate discovery or utility tools.
The tool covers the core decision operations: choice, score, noul, box, and batching independent questions. Minor gaps exist around endpoint discovery, health checks, or model metadata, but the main decision workflow is complete.
Calibrated judgments for text: yes/no probabilities, picks from your options, or scores.
Typed-question decisions from Jev: classify, moderate, route models and guard tool calls.
Deterministic contextual decision arbitration and action routing for autonomous software. Takes current state, context, or intent plus caller-supplied candidate actions, state transitions, routes, refusals, escalations, tools, or models and returns a deterministic ordered candidate field. Also provides persistent machine representations for memory, retrieval, indexing, and downstream coherence measurement.
Typed AI decisions for agents: classify, route, screen evidence, score and batch-label text.
