@cognitive-engine/orchestrator

Master cognitive pipeline for cognitive-engine. Orchestrates all modules into a single process() call.

Install

npm install @cognitive-engine/orchestrator

Usage

Full Pipeline

import { CognitiveOrchestrator } from '@cognitive-engine/orchestrator'
import { OpenAiLlmProvider, OpenAiEmbeddingProvider } from '@cognitive-engine/provider-openai'
import { MemoryStore } from '@cognitive-engine/store-memory'

const engine = new CognitiveOrchestrator({
  llm: new OpenAiLlmProvider({ apiKey: process.env.OPENAI_API_KEY, model: 'gpt-4o-mini' }),
  embedding: new OpenAiEmbeddingProvider({ apiKey: process.env.OPENAI_API_KEY }),
  store: new MemoryStore(),
})

const result = await engine.process('user-123', 'I feel stuck on this project')

result.percept.emotionalTone       // 'frustrated'
result.reasoning.intentions        // [{ type: 'empathize', priority: 10 }]
result.emotional.valence           // -0.3
result.social.rapport              // 0.65
result.metacognition.confidence    // 0.72
result.suggestedResponse           // AI-generated empathetic response

Selective Modules

Enable only what you need — disabled modules have zero overhead:

const engine = new CognitiveOrchestrator({
  llm, embedding, store,
  modules: {
    perception: true,
    memory: true,
    emotional: true,
    // reasoning, social, planning, etc. — all off
  },
})

With Events

import { CognitiveEventEmitter } from '@cognitive-engine/core'

const events = new CognitiveEventEmitter()
events.on('perception:complete', (percept) => analytics.track(percept))
events.on('episode:created', (episode) => console.log('Remembered:', episode.summary))

const engine = new CognitiveOrchestrator({ llm, embedding, store, events })

Pipeline

Message → Perception → [Memory + Reasoning] → [Emotional + Social + Mind + Plan + Temporal + Bandit] → Metacognition → Response

Modules in brackets run in parallel. The orchestrator manages data flow between stages.

License

Apache-2.0