Interface: ReasoningStageDto

@kortexya/reasoninglayer

---

@kortexya/reasoninglayer / Conversation / ReasoningStageDto

Interface: ReasoningStageDto

Defined in: src/types/conversation.ts:149

One executed stage of the conversation pipeline (LLM-as-sensor trace).

Properties

confidence?

optional confidence: number

Defined in: src/types/conversation.ts:161

Optional confidence for this stage (0.0–1.0).

---

detail?

optional detail: string

Defined in: src/types/conversation.ts:169

Free-text detail for display.

---

kind

kind: string

Defined in: src/types/conversation.ts:155

“sensor” (LLM was consulted) or “symbolic” (deterministic).

---

llmUsed

llmUsed: boolean

Defined in: src/types/conversation.ts:159

Whether the LLM was actually invoked at this stage.

---

matchedEntities

matchedEntities: MatchedEntityDto[]

Defined in: src/types/conversation.ts:163

Entities matched at this stage (sort + term id).

---

name

name: string

Defined in: src/types/conversation.ts:151

Stage id, e.g. “classify_intent”.

---

order

order: number

Defined in: src/types/conversation.ts:153

Execution order.

---

osfql?

optional osfql: string

Defined in: src/types/conversation.ts:165

OSFQL executed/produced at this stage, if any.

---

sensorPoint?

optional sensorPoint: string

Defined in: src/types/conversation.ts:157

For sensor stages: the sensor point consulted.

---

strategy?

optional strategy: string

Defined in: src/types/conversation.ts:167

Symbolic strategy used, if any.