Documentation

Everything you need to integrate Slaash into your agent, API, or workflow.

Quickstart

Send your first request in under 60 seconds. URL in, answer out. Python, JavaScript, and cURL examples.

Getting started

MCP Integration

Connect Slaash to Claude Desktop, Cursor, or VS Code. 12 agent primitives via MCP protocol. API key required.

MCP Server

Slaash Guide

How Slaash works: goal expansion, CRFR resonance fields, causal learning, feedback loops, output formats.

Core concepts

REST API

Full endpoint reference for all 12 primitives: extract, search, crawl, act, plan, stream, diff, render, and more.

API Reference

Trust Shield

How prompt injection detection works. 40+ attack patterns, trust levels, zero-width detection, multilingual.

Security

Self-Hosting Coming soon

Run Slaash as a single binary. Docker, Render, or bare metal deployment.

Deployment

Quick example

# Extract answer from any URL — 99% token reduction curl -X POST https://api.slaash.ai/v1/extract \ -H "Authorization: Bearer sk-..." \ -H "Content-Type: application/json" \ -d '{ "url": "https://en.wikipedia.org/wiki/Rust_(programming_language)", "goal": "when was Rust first released and who created it" }'
# Teach Slaash which nodes had the answer — improves future queries curl -X POST https://api.slaash.ai/v1/learn \ -H "Authorization: Bearer sk-..." \ -H "Content-Type: application/json" \ -d '{ "url": "https://en.wikipedia.org/wiki/Rust_(programming_language)", "goal": "when was Rust first released and who created it", "node_ids": [142, 289] }'