SDKs¶

Official client libraries for Sekha.

Available SDKs¶

Python SDK¶

Official Python client with full API coverage:

Features: - ✅ Full REST API coverage - ✅ Type hints & autocompletion - ✅ Async/await support - ✅ Pydantic models - ✅ 100% test coverage

Installation:

pip install sekha-sdk  # Coming soon to PyPI

JavaScript/TypeScript SDK¶

Official Node.js and browser client:

JavaScript SDK Documentation →

Features: - ✅ Full TypeScript support - ✅ Promise-based API - ✅ Browser and Node.js compatible - ✅ Auto-generated types - ✅ ESM and CommonJS

Installation:

npm install @sekha/sdk  # Coming soon to npm

Quick Examples¶

Python¶

from sekha import SekhaClient

client = SekhaClient(
    base_url="http://localhost:8080",
    api_key="your-api-key"
)

# Store conversation
convo = client.conversations.create(
    label="My Conversation",
    messages=[
        {"role": "user", "content": "Hello!"},
        {"role": "assistant", "content": "Hi there!"}
    ]
)

# Semantic search
results = client.query(
    query="What did we discuss?",
    limit=5
)

# Get context for LLM
context = client.context.assemble(
    query="Continue our chat",
    context_budget=8000
)

JavaScript/TypeScript¶

import { SekhaClient } from '@sekha/sdk';

const client = new SekhaClient({
  baseUrl: 'http://localhost:8080',
  apiKey: 'your-api-key'
});

// Store conversation
const convo = await client.conversations.create({
  label: 'My Conversation',
  messages: [
    { role: 'user', content: 'Hello!' },
    { role: 'assistant', content: 'Hi there!' }
  ]
});

// Semantic search
const results = await client.query({
  query: 'What did we discuss?',
  limit: 5
});

// Get context for LLM
const context = await client.context.assemble({
  query: 'Continue our chat',
  contextBudget: 8000
});

Coming Soon¶

Go SDK¶

Status: 🚧 Planned for Q2 2026

import "github.com/sekha-ai/sekha-go"

client := sekha.NewClient("http://localhost:8080", "api-key")

Rust SDK¶

Status: 🚧 Planned for Q2 2026

use sekha::Client;

let client = Client::new("http://localhost:8080", "api-key");

SDK Code Examples¶

Comprehensive examples for common use cases:

Storing Conversations¶

# Python example
client.conversations.create(
    label="Project Meeting",
    messages=[
        {"role": "user", "content": "Let's discuss the roadmap"},
        {"role": "assistant", "content": "Sure! What aspects?"}  
    ],
    folder="work/project-a",
    importance=8
)

Semantic Search¶

// TypeScript example
const results = await client.query({
  query: "architecture decisions",
  folder: "work",
  limit: 10,
  threshold: 0.7
});

for (const result of results.results) {
  console.log(`${result.relevance_score}: ${result.message}`);
}

Context Assembly¶

# Get optimal context for next LLM call
context = client.context.assemble(
    query="Continue discussing the payment system",
    context_budget=8000,
    include_summaries=True
)

# Use with OpenAI
import openai
response = openai.ChatCompletion.create(
    model="gpt-4",
    messages=[
        {"role": "system", "content": context.context},
        {"role": "user", "content": "What were our next steps?"}
    ]
)

SDK Development¶

Contribute to SDK development:

Next Steps¶

Python SDK - Full Python documentation
JavaScript SDK - Full JS/TS documentation
API Reference - REST API docs
Guides - Use case examples