Encrypted, verifiable, decentralised. Every memory is SHA-256 hashed, stored on Shelby Protocol, and anchored on Aptos.
npm install @forestinfra/shelmempip install shelmem
The memory layer that proves memories are real.
SHA-256 content hash on every write. On recall, content is re-downloaded and the hash verified. Tampered memories are flagged instantly.
End-to-end encryption. Memories are encrypted before upload to Shelby. Key derived from your Aptos private key — zero additional secrets.
pgvector embeddings stored alongside memories. Search by meaning — not just exact keyword matching.
Every memory write submits an Aptos transaction. Cryptographic proof that the memory existed at that exact moment.
Categorise memories as facts, decisions, preferences, or observations. Filter on recall by type.
Content lives on Shelby Protocol's distributed hot storage. No single point of failure.
Drop-in integrations for LangChain, CrewAI, Vercel AI SDK, and Coinbase AgentKit.
Published on npm and PyPI. Same API, same verification, same encryption.
Write with proof. Recall with tamper detection. Search by meaning.
import { ShelMem, openaiEmbeddings } from '@forestinfra/shelmem';
const mem = new ShelMem({
supabaseUrl, supabaseKey,
encrypt: true,
embeddingProvider: openaiEmbeddings(OPENAI_KEY),
});
// Write — encrypted, hashed, stored on Shelby, anchored on Aptos
await mem.write('agent-01', 'Bought ETH at $2,847', 'analysis', 'decision');
// Recall — decrypted, hash verified
const memories = await mem.recall('agent-01');
// → [{ memory, verified: true, memory_type: 'decision' }]
// Semantic search — find by meaning
const results = await mem.search('what do I know about ETH?');
// → [{ memory_preview, similarity: 0.89 }]Works With
Vercel AIInstall the SDK and write your first verified memory in under a minute.