fleet-memory/hindsight-integrations/ai-sdk
Nicolò Boschi b6a4f17cbe
fix: resolve all Dependabot security alerts (#702)
- requests: bump minimum to >=2.33.0 (CVE temp file reuse)
- streamlit: bump minimum to >=1.54.0 (SSRF/NTLM exposure)
- picomatch: add npm override for >=2.3.2/<3 || >=4.0.4 (ReDoS + method injection)
- flatted: tighten override to >=3.4.2 (prototype pollution)
- yaml: add npm override for >=1.10.3 (stack overflow)
- rustls-webpki: cargo update to 0.103.10 (CRL distribution point)
- Also fix pre-existing ty lint error in metrics.py (type: ignore for Windows resource import)
- Pygments: no patch available (<=2.19.2 vulnerable, no fix released)
2026-03-26 13:15:36 +01:00
..
src feat: improve ai sdk tools (#394) 2026-02-18 13:06:03 +01:00
tests feat(typescript-client): Deno compatibility (#607) 2026-03-18 14:25:35 +01:00
.gitignore feat: ai sdk integration (#299) 2026-02-04 20:25:59 +01:00
deno.json feat(typescript-client): Deno compatibility (#607) 2026-03-18 14:25:35 +01:00
deno.lock feat(typescript-client): Deno compatibility (#607) 2026-03-18 14:25:35 +01:00
package-lock.json fix: resolve all Dependabot security alerts (#702) 2026-03-26 13:15:36 +01:00
package.json fix: resolve all Dependabot security alerts (#702) 2026-03-26 13:15:36 +01:00
README.md docs: add AI SDK integration documentation (#304) 2026-02-05 17:05:18 +01:00
tsconfig.json feat: ai sdk integration (#299) 2026-02-04 20:25:59 +01:00
tsup.config.ts feat(typescript-client): Deno compatibility (#607) 2026-03-18 14:25:35 +01:00
vitest.config.ts feat: ai sdk integration (#299) 2026-02-04 20:25:59 +01:00

Hindsight Memory Integration for Vercel AI SDK

Give your AI agents persistent, human-like memory using Hindsight with the Vercel AI SDK.

Quick Start

npm install @vectorize-io/hindsight-ai-sdk @vectorize-io/hindsight-client ai zod
import { HindsightClient } from '@vectorize-io/hindsight-client';
import { createHindsightTools } from '@vectorize-io/hindsight-ai-sdk';
import { generateText } from 'ai';
import { anthropic } from '@ai-sdk/anthropic';

// 1. Initialize Hindsight client
const hindsightClient = new HindsightClient({
  apiUrl: 'http://localhost:8000',
});

// 2. Create memory tools
const tools = createHindsightTools({ client: hindsightClient });

// 3. Use with AI SDK
const result = await generateText({
  model: anthropic('claude-sonnet-4-20250514'),
  tools,
  system: `You have long-term memory. Use:
  - 'recall' to search past conversations
  - 'retain' to remember important information
  - 'reflect' to synthesize insights from memories`,
  prompt: 'Remember that Alice loves hiking and prefers spicy food',
});

console.log(result.text);

Features

Three Memory Tools: retain (store), recall (retrieve), and reflect (reason over memories) AI SDK 6 Native: Works with generateText, streamText, and ToolLoopAgent Multi-User Support: Dynamic bank IDs per call for multi-user scenarios Type-Safe: Full TypeScript support with Zod schemas Flexible Client: Works with the official TypeScript client or custom HTTP clients

Documentation

📖 Full Documentation

The complete documentation includes:

  • Detailed tool descriptions and parameters
  • Advanced usage patterns (streaming, multi-user, ToolLoopAgent)
  • HTTP client example (no dependencies)
  • TypeScript types and API reference
  • Best practices and system prompt examples

Running Hindsight Locally

# Install and run with embedded mode (no setup required)
uvx hindsight-embed@latest -p myapp daemon start

# The API will be available at http://localhost:8000

Examples

Full examples are available in the GitHub repository.

Support

License

MIT