Classify messages, rate content, or estimate probabilities using defined question types. Receive machine-readable answers, with confidence scores for classification and ratings, and apply the same questions to multiple inputs in a batch.
中文简介正在准备中。当前显示英语简介。
Typed decision expert. ALWAYS invoke this skill when you need structured classification, scoring, or probability estimation — routing tickets, grading content, detecting intent, or any task requiring typed answers with calibrated confidence.
安装前可以查看交给 AI 代理的指令内容。
Use these tools when you need structured, typed decisions rather than free-form LLM text. Decision models return machine-readable answers with probabilities and confidence scores.
Send a state (context) and one or more typed questions to a decision model.
| Type | Purpose | Criteria | Answer Fields |
|---|---|---|---|
choice | Classify into one of N options | {option: description} map | choice, probabilities, confidence |
score | Rate on a scale | [level_labels] array | score, legend, probabilities, confidence |
noul | Estimate a probability (0-1) | (none) | noul |
tai tool decision_decide --state "Customer says: my billing is wrong, I was charged twice" --questions '{"department":{"type":"choice","instructions":"Route to the right department","criteria":{"billing":"Payment issues","technical":"Product bugs","sales":"Purchase inquiries"}},"is_urgent":{"type":"noul","instructions":"Is this urgent?"}}'
tai tool decision_decide --state "The agent resolved the issue in 2 minutes with a clear explanation" --questions '{"quality":{"type":"score","instructions":"Rate the support quality","criteria":["Poor","Below average","Average","Good","Excellent"]}}'
tai tool decision_decide --state "some context" --questions '{"q1":{"type":"noul","instructions":"Is this positive?"}}' --provider t862699292876.typesafe
tai tool decision_decide --state "some context" --questions '{"q1":{"type":"noul","instructions":"Is this positive?"}}' --model jev-latest
tai tool decision_decide --states '["Customer A says billing is wrong","Customer B loves the product","Customer C wants a refund"]' --questions '{"sentiment":{"type":"choice","instructions":"Classify sentiment","criteria":{"positive":"Happy","negative":"Unhappy","neutral":"Neither"}}}'
tai tool decision_decide --state_file customer_message.txt --questions_file routing_questions.json
| Parameter | Type | Required | Description |
|---|---|---|---|
| state | string/object/array | yes* | The context for the decision (*mutually exclusive with states) |
| questions | JSON object | yes | Map of question_id to question definition |
| model | string | no | Model name override (default: provider's configured model) |
| provider | string | no | Provider connector ID (default: decision role or first available) |
| timeout | integer | no | Timeout in seconds (default: 60) |
| states | array | no* | Array of state items for batch decisions (*mutually exclusive with state) |
| state_file | string | no | Read state from file (tai CLI only) |
| questions_file | string | no | Read questions from JSON file (tai CLI only) |
Single state:
{
"answers": {
"department": {
"type": "choice",
"choice": "billing",
"probabilities": {"billing": 0.92, "technical": 0.05, "sales": 0.03},
"confidence": 0.89
},
"is_urgent": {
"type": "noul",
"noul": 0.73
}
},
"model": "jev-1.13.0",
"usage": {"input_tokens": 245, "output_tokens": 38}
}
Note:
modelin the response is the resolved concrete model ID (e.g.jev-1.13.0), not the alias you requested (e.g.jev-latest).
Batch states:
{
"results": [
{"state_index": 0, "answers": {...}, "model": "jev-1.13.0", "usage": {...}},
{"state_index": 1, "answers": {...}, "model": "jev-1.13.0", "usage": {...}},
{"state_index": 2, "error": "timed out after 60s"}
]
}
List available decision providers and their models.
tai tool decision_providers
| Parameter | Type | Required | Description |
|---|---|---|---|
| capability | string | no | Filter by capability (default: decision) |
Returns a list of providers with available models, connector IDs, and availability status.
state and states are mutually exclusive; provide one or the other.type (choice/score/noul) and instructions.criteria is required for choice (map) and score (array); noul must not have criteria.timeout defaults to 60s; values ≤ 0 are treated as the default.model is the resolved concrete model ID (e.g. jev-1.13.0), not the requested alias.还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Manage agents on a host by listing their capabilities, reading source code, and downloading or deploying agents in the smith namespace. Check available language model connections and their roles without exposing API keys.
中文简介正在准备中。当前显示英语简介。
Turn recorded speech into text from supported audio files such as mp3, m4a, and wav. Detect the language automatically or specify it, and list available speech recognition providers and models before choosing one.
中文简介正在准备中。当前显示英语简介。
Check kanban boards, which organize tasks into columns, and retrieve task lists filtered by status, assistant, or board. Use the returned data to see task counts and review work in progress.
中文简介正在准备中。当前显示英语简介。
Start and manage long-running background services such as development servers. Check their status and output logs, verify listening ports, retrieve local URLs, and restart services while keeping the same service ID.
中文简介正在准备中。当前显示英语简介。
Look up callable processes in Yao, inspect their arguments and return types, and check whether a process name is valid. Use the documentation to prepare calls and find suggested alternatives for misspelled names.
中文简介正在准备中。当前显示英语简介。
Describe photos, screenshots, and charts in text, generate images from written prompts, or edit existing images. Supports background replacement, transparent output, and edits limited to an area defined by a mask.
中文简介正在准备中。当前显示英语简介。