Ready-to-run example projects demonstrating SerialMemory integration patterns.
-
Start SerialMemory locally:
./dev-bootstrap.sh # Linux/Mac .\dev-bootstrap.ps1 # Windows
-
Run an example:
# .NET cd examples/ai-second-brain/dotnet dotnet run # Node.js cd examples/ai-second-brain/node npm install && npm start
A persistent "second brain" for AI assistants. Demonstrates:
- Storing notes, decisions, and learnings
- Semantic search for relevant context
- User persona management
- Multi-hop knowledge graph traversal
Use cases: Personal AI assistant, knowledge management, decision logging
Project-specific memory isolation. Demonstrates:
- Scoped memory per project/repository
- Metadata-based filtering
- Cross-project search when needed
- Multi-tenant patterns
Use cases: IDE integration, development tools, multi-repo AI assistants
Lower-level SDK usage examples are in the sdks/ directory:
Well-structured content improves search relevance:
[Category]: [Brief title]
Context:
[Background information]
Details:
[Main content]
Tags: tag1, tag2
Date: YYYY-MM-DD
await client.ingest(content, {
source: 'my-app',
metadata: {
type: 'decision',
project: 'acme',
importance: 'high',
tags: ['architecture', 'database']
}
});try {
const result = await client.search(query);
} catch (error) {
if (error instanceof RateLimitExceededError) {
await sleep(error.retryAfter);
// retry
} else if (error instanceof UsageLimitExceededError) {
console.log('Upgrade plan or wait for reset');
}
}All examples support these environment variables:
| Variable | Default | Description |
|---|---|---|
SERIALMEMORY_URL |
http://localhost:5000 |
API base URL |
SERIALMEMORY_API_KEY |
demo-key |
API key for authentication |