Memory infrastructure for AI agents. Your Qdrant, your hardware, our intelligence.

PyPI · Dashboard · API Docs · Community Edition

npm install engrammemory-ai

Quick Start

import { Engram } from 'engrammemory-ai';

const engram = new Engram({
  apiKey: 'eng_live_xxx',
  qdrantUrl: 'http://localhost:6333'
});

// Store — embedded & deduplicated by Engram, stored in YOUR Qdrant
await engram.store('User prefers TypeScript and dark mode', {
  category: 'preference'
});

// Search — three-tier recall (hot → hash → vector)
const results = await engram.search('What does the user prefer?');
for (const r of results) {
  console.log(`[${r.tier}] ${r.content} (${r.score.toFixed(2)})`);
}

// Forget
await engram.forget('mem_abc123');

CommonJS:

const { Engram } = require('engrammemory-ai');

Why Engram?

| | Engram | Mem0 | Supermemory | Zep | |---|---|---|---|---| | Data ownership | Your Qdrant, your hardware | Their cloud | Their cloud | Their cloud | | Three-tier recall | Hot → Hash → Vector | Single-tier | Single-tier | Two-tier | | Stateless intelligence | Embeds + classifies, never stores | Stores everything | Stores everything | Stores everything | | Edge/fleet support | Per-device agents, zone isolation | No | No | Limited | | Local-first + overflow | Store locally, overflow to cloud | Cloud only | Cloud only | Cloud only | | Open community edition | OpenClaw skill, full recall engine | No | Partial | No |

Three-Tier Recall

Every search flows through three tiers automatically:

const results = await engram.search('database config');
for (const r of results) {
  console.log(`[${r.tier}] ${r.content} (${r.score.toFixed(2)})`);
  // [hot]    PostgreSQL 15 with read replicas (0.94)   — sub-ms, cached
  // [hash]   Migrated from MongoDB last quarter (0.81)  — O(1) LSH lookup
  // [vector] Old DB credentials in vault (0.67)         — full ANN fallback
}

Auto-Recall for Agent Prompts

// Inject relevant memories into your agent's system prompt
const memories = await engram.recall('User is asking about their database setup');
const systemPrompt = `You know: ${memories.map(m => m.content).join(', ')}`;

Multi-Agent / Fleet

const engram = new Engram({
  apiKey: 'eng_live_xxx',
  project: 'icu-floor-3'
});

// Store with agent tracking
await engram.store('Patient allergic to penicillin', {
  category: 'fact',
  agent: 'tablet-icu-3a'
});

// Search scoped to project
const results = await engram.search('allergies', {
  agent: 'tablet-icu-3a'
});

API Reference

Constructor

new Engram({
  apiKey: string;                          // Your Engram API key
  qdrantUrl?: string;                      // Default: "http://localhost:6333"
  baseUrl?: string;                        // Default: "https://api.engrammemory.ai"
  collection?: string;                     // Default: "agent-memory"
  project?: string;                        // Memory isolation namespace
  maxRetries?: number;                     // Default: 3
  timeout?: number;                        // Default: 30000 (ms)
})

Methods

| Method | Returns | Description | |---|---|---| | store(content, options?) | Promise<StoreResult> | Store a memory (embedded + deduplicated + compressed) | | search(query, options?) | Promise<SearchResult[]> | Three-tier recall: hot → hash → vector | | recall(context, options?) | Promise<SearchResult[]> | Auto-inject relevant memories for agent prompts | | get(memoryId) | Promise<Memory> | Get a specific memory by ID | | forget(memoryId) | Promise<boolean> | Remove from all three tiers | | forgetByQuery(query) | Promise<number> | Forget memories matching a search, returns count | | list(options?) | Promise<Memory[]> | List memories with filtering | | health() | Promise<HealthStatus> | Check Engram + Qdrant connectivity |

Types

import type { Memory, SearchResult, StoreResult, HealthStatus } from 'engrammemory-ai';

| Type | Fields | |---|---| | Memory | id, content, category, metadata, createdAt, accessCount | | SearchResult | id, content, score, tier ("hot" | "hash" | "vector"), category, metadata | | StoreResult | id, stored, deduplicated | | HealthStatus | status, qdrantConnected, version |

Exceptions

import { EngramError, AuthenticationError, RateLimitError, NotFoundError, ValidationError } from 'engrammemory-ai';

| Exception | HTTP Code | When | |---|---|---| | AuthenticationError | 401 | Invalid or missing API key | | RateLimitError | 429 | Too many requests (auto-retries with Retry-After) | | NotFoundError | 404 | Memory ID doesn't exist | | ValidationError | 400/422 | Invalid request body | | EngramError | Other | Base class for all errors |

Community Edition

Don't need cloud? The community edition runs the full three-tier recall engine locally as an OpenClaw skill with no API dependency.

Links

• Python SDK — pip install engrammemory-ai
• Dashboard
• API Docs
• GitHub
• Community Edition