@sisu-ai/vector-chroma

Chroma vector-store adapter for Sisu RAG composition.

Exports

• createChromaVectorStore

The created adapter exposes upsert, query, and delete, and is intended to be injected into backend-agnostic RAG tools such as @sisu-ai/tool-rag.

For reusable chunking, record preparation, and app-side seeding flows, pair this adapter with @sisu-ai/rag-core.

Philosophy

@sisu-ai/vector-chroma is a backend adapter package.

• It should only know how to talk to Chroma.
• It should not own agent-facing tool behavior.
• It should not own generic RAG chunking or prompt logic.

That separation keeps backend swaps straightforward later.

How It Fits In The Stack

Typical composition looks like this:

• @sisu-ai/vector-core defines the VectorStore contract
• @sisu-ai/vector-chroma implements the contract for Chroma
• @sisu-ai/rag-core uses the contract for reusable ingestion/retrieval mechanics
• @sisu-ai/tool-rag uses rag-core to expose model-facing tools
• @sisu-ai/mw-rag can use the same VectorStore directly in middleware flows

Typical Usage

With @sisu-ai/rag-core

import { storeRagContent } from '@sisu-ai/rag-core';
import { createChromaVectorStore } from '@sisu-ai/vector-chroma';

const vectorStore = createChromaVectorStore({
  chromaUrl: process.env.CHROMA_URL,
  namespace: 'docs',
});

await storeRagContent({
  content: 'Sisu uses small, composable packages.',
  embeddings,
  vectorStore,
});

With @sisu-ai/tool-rag

import { createRagTools } from '@sisu-ai/tool-rag';
import { createChromaVectorStore } from '@sisu-ai/vector-chroma';

const ragTools = createRagTools({
  embeddings,
  vectorStore: createChromaVectorStore({ namespace: 'docs' }),
});

With @sisu-ai/mw-rag

import { ragIngest, ragRetrieve } from '@sisu-ai/mw-rag';
import { createChromaVectorStore } from '@sisu-ai/vector-chroma';

const vectorStore = createChromaVectorStore({ namespace: 'docs' });

agent
  .use(ragIngest({ vectorStore }))
  .use(ragRetrieve({ vectorStore, topK: 4 }));

Using This As The Template For New Providers

If you add another backend, the pattern is:

1. create a new vector-* package
2. implement the VectorStore contract from @sisu-ai/vector-core
3. keep backend SDK details inside that package only
4. reuse it from rag-core, tool-rag, and mw-rag

@sisu-ai/vector-chroma is the example implementation of that pattern.

Contributing

We build Sisu in the open. Contributions welcome.

Contributing Guide · Report a Bug · Request a Feature · Code of Conduct

Star on GitHub if Sisu helps you build better agents.

Quiet, determined, relentlessly useful.

Apache 2.0 License