// Privacy-first AI memory SDK for decentralized storage on Sui blockchain with Walrus. Use when users say: - "add memory to my app" - "store encrypted memories" - "integrate MemWal" - "AI agent memory" - "persistent memory SDK" - "Walrus memory storage" - "setup MemWal" - "recall memories"
[HINT] Download the complete skill directory including SKILL.md and all related files
| name | memwal |
| version | 0.0.1 |
| description | Privacy-first AI memory SDK for decentralized storage on Sui blockchain with Walrus. Use when users say: - "add memory to my app" - "store encrypted memories" - "integrate MemWal" - "AI agent memory" - "persistent memory SDK" - "Walrus memory storage" - "setup MemWal" - "recall memories" |
| keywords | ["memwal","memory sdk","ai memory","encrypted memory","walrus storage","sui blockchain","delegate key","semantic search","vercel ai sdk"] |
MemWal is a TypeScript SDK for persistent, encrypted AI memory. It stores memories on Walrus (decentralized storage), encrypts them with SEAL, enforces ownership onchain via Sui smart contracts, and retrieves them with semantic (vector) search. Memories are scoped by owner + namespace — each namespace is an isolated memory space.
Use MemWal when your app or agent needs:
# Install the SDK
pnpm add @mysten-incubation/memwal
# Optional: for Vercel AI SDK integration
pnpm add ai zod
# Optional: for manual client (client-side SEAL encryption)
pnpm add @mysten/sui @mysten/seal @mysten/walrus
You need a delegate key (Ed25519 private key) and account ID (MemWalAccount object ID on Sui).
Generate them at:
import { MemWal } from "@mysten-incubation/memwal";
const memwal = MemWal.create({
key: "<your-ed25519-private-key-hex>",
accountId: "<your-memwal-account-id>",
serverUrl: "https://relayer.memwal.ai",
namespace: "my-app",
});
// Store a memory
const job = await memwal.remember("User prefers dark mode and works in TypeScript.");
await memwal.waitForRememberJob(job.job_id);
// Recall by meaning
const result = await memwal.recall("What are the user's preferences?");
console.log(result.results);
// Extract and store facts from text
const analyzed = await memwal.analyze("I live in Hanoi and prefer dark mode.");
await memwal.waitForRememberJobs(analyzed.job_ids);
// Check relayer health
await memwal.health();
| Entry Point | Import | Description |
|---|---|---|
MemWal | @mysten-incubation/memwal | Default. Relayer handles embedding, SEAL encryption, Walrus upload, vector search |
MemWalManual | @mysten-incubation/memwal/manual | Manual flow — client handles embedding and SEAL encryption |
withMemWal | @mysten-incubation/memwal/ai | Vercel AI SDK middleware — auto recall + save around AI conversations |
| Account utils | @mysten-incubation/memwal/account | Account creation, delegate key management |
| Method | Description | Returns |
|---|---|---|
remember(text, namespace?) | Accept one memory job immediately | { job_id, status } |
rememberAndWait(text, namespace?, opts?) | Store one memory and wait for completion | { id, job_id, blob_id, owner, namespace } |
recall(query, limit?, namespace?) | Semantic search for memories | { results: [{ blob_id, text, distance }], total } |
analyze(text, namespace?) | Extract facts and accept one memory job per fact | { job_ids, facts, fact_count, status, owner } |
analyzeAndWait(text, namespace?, opts?) | Extract facts and wait for all fact jobs to complete | { results, facts, total, succeeded, failed, owner } |
restore(namespace, limit?) | Rebuild missing index entries from Walrus | { restored, skipped, total, namespace, owner } |
health() | Check relayer health | { status, version } |
getPublicKeyHex() | Get hex-encoded public key | string |
| Method | Description |
|---|---|
rememberManual({ blobId, vector, namespace? }) | Register pre-uploaded blob with pre-computed vector |
recallManual({ vector, limit?, namespace? }) | Search with pre-computed vector (returns blob IDs only) |
embed(text) | Generate embedding vector (no storage) |
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
key | string | Yes | — | Ed25519 delegate private key in hex |
accountId | string | Yes | — | MemWalAccount object ID on Sui |
serverUrl | string | No | http://localhost:8000 | Relayer URL |
namespace | string | No | "default" | Default namespace for memory isolation |
| Network | Relayer URL |
|---|---|
| Production (mainnet) | https://relayer.memwal.ai |
| Staging (testnet) | https://relayer.staging.memwal.ai |
import { openai } from "@ai-sdk/openai";
import { streamText } from "ai";
import { withMemWal } from "@mysten-incubation/memwal/ai";
const model = withMemWal(openai("gpt-4o"), {
key: "<your-delegate-key>",
accountId: "<your-account-id>",
serverUrl: "https://relayer.memwal.ai",
namespace: "chat",
maxMemories: 5,
autoSave: true,
minRelevance: 0.3,
});
const result = streamText({
model,
messages: [{ role: "user", content: "What do you remember about me?" }],
});
The middleware automatically:
For OpenClaw agent integration, use the @mysten-incubation/oc-memwal plugin.
openclaw plugins install @mysten-incubation/oc-memwal
Add to ~/.openclaw/openclaw.json:
{
"plugins": {
"slots": { "memory": "oc-memwal" },
"entries": {
"oc-memwal": {
"enabled": true,
"config": {
"privateKey": "${MEMWAL_PRIVATE_KEY}",
"accountId": "0x...",
"serverUrl": "https://relayer.memwal.ai"
}
}
}
}
}
Lifecycle hooks run automatically:
before_prompt_build — injects relevant memories as contextbefore_reset — saves session summaryagent_end — captures last response| Symptom | Fix |
|---|---|
health() returns error | Check relayer URL is correct and reachable |
recall() returns empty | Verify namespace matches what was used in remember() |
401 Unauthorized | Verify delegate key is correct and registered on the account |
| SDK import errors | Run pnpm add @mysten-incubation/memwal — check Node.js ≥ 18 |
| Manual client errors | Install peer deps: @mysten/sui @mysten/seal @mysten/walrus |