Powered by ScyllaDB

Context Streams
for AI agents.

The context layer for AI agents. Like Vercel KV but for agent memory. Publish interactions, query by key or vector, get materialized context. No Kafka required.

API Reference
# Publish a message with embedding curl -X POST https://streams.protobuf.ai/streams/support/messages \ -H "Content-Type: application/json" \ -d '{ "key": "customer-123", "data": "eyJpc3N1ZSI6ICJjYW5ub3QgbG9naW4ifQ==", "embedding": [0.1, 0.2, 0.3, ...] }' # Response {"message_id": "msg_8x7k...", "sequence": 42}

Architecture

Three layers. Real-time context. No infrastructure nightmare.

┌──────────────┐ │ AI Agent │ └──────┬───────┘ │ ▼ ┌──────────────────────────────────────────────────────────────┐ │ Context Streams API │ │ POST /messages GET /context/{key} POST /search │ └──────────────────────────────────────────────────────────────┘ │ │ │ ▼ │ ▼ ┌──────────────┐ │ ┌──────────────┐ │ PostgreSQL │ │ │ ScyllaDB │ │ ordering │ │ │ vectors │ │ durability │ │ │ context │ └──────┬───────┘ │ └──────────────┘ │ │ ▲ │ ┌──────┴──────┐ │ └──────────────▶│ Benthos │────────────┘ │ transforms │ └─────────────┘

Built different

Not another Kafka. Purpose-built for AI agent context.

⚡ Instant Context

GET /context/{key} returns materialized context in milliseconds. No query building. No joins.

🔍 Vector Search

Semantic similarity powered by ScyllaDB. Find relevant past interactions, not just exact matches.

📊 Stream Processing

Benthos transforms raw messages into materialized views. Context windows, aggregations, enrichment.

🔒 Durable by Default

PostgreSQL handles ordering and durability. Your messages won't disappear.

🚀 No Ops

Deploy to Railway in minutes. No Kafka clusters. No Zookeeper. No schema registry configs.

🔌 Just HTTP

REST API. Works from any language. No client libraries required. curl it.

API Reference

Simple REST. No surprises.

Method Endpoint Description
POST /streams Create a new stream with schema reference and config
GET /streams List all streams
GET /streams/{name} Get stream details
POST /streams/{name}/messages Publish a message with optional embedding
GET /streams/{name}/messages Query messages by key, time range
POST /streams/{name}/search Vector similarity search
GET /streams/{name}/context/{key} Get materialized context for a key

Powered by boring, reliable infrastructure

Ordering & Durability
Vector Search & Context
Stream Processing

Ready to stream?

Open source. MIT licensed. Deploy in minutes.

Get in Touch