Tutors that remember where the student got stuck

1. 01 · Ingest

   Scope what you ingest

   Every document lands with a groupName: the sets it belongs to. Those sets are what retrieval intersects later, so the structure you choose here is the precision you get there.

   context.add

   import AlchemystAI from "@alchemystai/sdk";const client = new AlchemystAI(); // reads ALCHEMYST_AI_API_KEYawait client.v1.context.add({  context_type: "resource",  scope: "internal",  source: "syllabus",  documents: [{    content: "Unit 4: Cell division. Mitosis produces two identical diploid cells. Meiosis produces four haploid gametes.",  }],  metadata: {    fileName: "bio-101-unit-4.pdf",    groupName: ["bio_101", "unit_4"],   // the sets this belongs to  },});

2. 02 · Write

   Write the misconception, not the chat

   What matters for next time is not the whole exchange. It is what the student believed, and what finally corrected it.

   context.memory.add

   await client.v1.context.memory.add({  sessionId: "student_3317",  contents: [    { role: "user", content: "Wait, so meiosis is the one that makes identical cells?" },    {      role: "assistant",      content: "Close, but that is mitosis. Meiosis makes four different cells with half the chromosomes.",    },  ],  metadata: { groupName: ["bio_101", "student_3317"] },});

3. 03 · Search

   Search before explaining

   Search intersects the scopes, subtracts superseded and duplicate content, and ranks what survives. Only that reaches the model, and the whole decision is recorded as a Context Trace.

   context.search

   const { contexts } = await client.v1.context.search({  query: "Explain why meiosis matters for genetic variation",  scope: "internal",  similarity_threshold: 0.8,  minimum_similarity_threshold: 0.5,  metadata: { groupName: ["bio_101", "unit_4"] },   // ∩ narrow scope});// − superseded, deduplicated → ranked → into the window// Every search is recorded as a Context Trace.const reply = await llm.respond(message, { context: contexts });

npm install @alchemystai/sdk or pip install alchemystai, both ship the same client. Full reference in the docs.