---

name: memwal version: 0.0.1 description: | Walrus Memory SDK — portable agent memory that works across apps, sessions, and workflows.

Use when users say:

• "add memory to my app"
• "portable agent memory"
• "integrate Walrus Memory"
• "AI agent memory"
• "memory across agents"
• "Walrus memory storage"
• "setup Walrus Memory"
• "recall memories"

keywords:

• memwal
• walrus memory
• memory sdk
• ai memory
• portable memory
• walrus storage
• sui blockchain
• delegate key
• semantic search
• vercel ai sdk

---

Walrus Memory — Portable Agent Memory

Walrus Memory enables AI agents to operate reliably across apps and sessions, without losing context. 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. Memory is portable by design — not tied to a single runtime or provider — and scoped by owner + namespace for isolation and coordination.

---

When to Use

Use Walrus Memory when your app or agent needs:

• Portable memory — persists outside prompts and context windows, moves across agents, apps, and workflows
• Full owner control — programmable permissions and explicit ownership define how memory is shared and accessed
• Agent coordination — shared memory spaces help agents coordinate across long-running and multi-step workflows
• Semantic recall — retrieve memories by meaning, not just keywords
• Verifiable integrity — memory integrity can be independently verified without centralized trust
• Cross-app memory — not tied to a single runtime or provider, share memory between apps via delegate keys

---

When NOT to Use

• Temporary conversation context that only matters in the current session
• Large file storage (Walrus Memory is optimized for text memories)
• Use cases that don't need encryption or decentralization

---

Installation

# 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

---

Quick Start

1. Get Your Credentials

You need a delegate key (Ed25519 private key) and account ID (Walrus Memory account object ID on Sui).

Generate them at:

• Production: https://memwal.ai or https://memwal.wal.app
• Staging: https://staging.memwal.ai

2. Initialize the SDK

import { MemWal } from "@mysten-incubation/memwal";

const memwal = MemWal.create({
  key: process.env.MEMWAL_PRIVATE_KEY!,
  accountId: process.env.MEMWAL_ACCOUNT_ID!,
  serverUrl: process.env.MEMWAL_SERVER_URL ?? "https://relayer.memwal.ai",
  namespace: "my-app",
});

3. Store and Recall Memories

// Store one already-distilled fact and wait until it is indexed.
await memwal.rememberAndWait(
  "User prefers dark mode and works in TypeScript.",
  undefined,
  { timeoutMs: 30_000 },
);

// Recall by meaning
const result = await memwal.recall({ query: "What are the user's preferences?" });
console.log(result.results);

// Extract facts from free-form text and wait until all accepted facts are indexed.
const analyzed = await memwal.analyzeAndWait(
  "I live in Hanoi and prefer dark mode.",
  undefined,
  { timeoutMs: 30_000 },
);
console.log(analyzed.facts.map((fact) => fact.text));

// Check relayer health
await memwal.health();

Use *AndWait when a workshop UI saves and then immediately recalls in the same flow. Indexing can lag by a few seconds, so remember() / analyze() may return before recall can find the new memory. Manual polling is still available for advanced async UIs:

const accepted = await memwal.remember("User likes Sui.");
const stored = await memwal.waitForRememberJob(accepted.job_id, {
  pollIntervalMs: 750,
  timeoutMs: 30_000,
});

---

SDK Entry Points

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

---

API Surface

Walrus Memory Methods

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?, maxDistance? }) (preferred) or 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

Lower-Level Methods

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)

All Response Shapes

interface RememberAcceptedResult {
  job_id: string;
  status: string;
}

interface RememberJobStatus {
  job_id: string;
  status: "pending" | "running" | "uploaded" | "done" | "failed" | "not_found";
  owner?: string;
  namespace?: string;
  blob_id?: string;
  error?: string;
}

interface RememberResult {
  id: string;
  job_id?: string;
  blob_id: string;
  owner: string;
  namespace: string;
}

interface RecallMemory {
  blob_id: string;
  text: string;
  distance: number;
}

interface RecallResult {
  results: RecallMemory[];
  total: number;
}

interface RecallOptions {
  limit?: number;
  topK?: number;
  namespace?: string;
  maxDistance?: number;
}

interface RememberBulkAcceptedResult {
  job_ids: string[];
  total: number;
  status: string;
}

interface AnalyzedFact {
  text: string;
  id: string;
  job_id?: string;
  blob_id?: string;
}

interface AnalyzeResult {
  job_ids: string[];
  facts: AnalyzedFact[];
  fact_count: number;
  status: string;
  owner: string;
}

interface RememberBulkStatusItem {
  job_id: string;
  status: "pending" | "running" | "uploaded" | "done" | "failed" | "not_found";
  blob_id?: string;
  error?: string;
}

interface RememberBulkStatusResult {
  results: RememberBulkStatusItem[];
}

interface RememberBulkItemResult {
  id: string;
  blob_id: string;
  status: "done" | "failed" | "timeout";
  namespace: string;
  error?: string;
}

interface RememberBulkResult {
  results: RememberBulkItemResult[];
  total: number;
  succeeded: number;
  failed: number;
}

interface AnalyzeWaitResult extends RememberBulkResult {
  facts: AnalyzedFact[];
  owner: string;
}

interface EmbedResult {
  vector: number[];
}

interface RestoreResult {
  restored: number;
  skipped: number;
  total: number;
  namespace: string;
  owner: string;
}

interface HealthResult {
  status: string;
  version: string;
  mode?: string;
  prompt_versions?: {
    extract: string;
    ask: string;
  };
  relayerVersion?: string;
  apiVersion?: string;
  minSupportedSdk?: {
    typescript: string;
    python: string;
    mcp: string;
  };
  featureFlags?: Record<string, boolean>;
  deprecations?: Array<{
    surface: string;
    deprecatedSince: string;
    removalApiVersion: string;
    guidance: string;
  }>;
  build?: {
    commit?: string;
    buildTimestamp?: string;
  };
}

facts[].text is the extracted fact text to render in UIs. job_ids[] aligns with the accepted fact jobs; use analyzeAndWait() when the UI needs those facts indexed before continuing.

Namespace Semantics

A namespace is an opaque, flat string label scoped to a single owner. It is the unit of memory isolation: a recall in namespace A will never surface entries written to namespace B, even for the same owner, and never surfaces other owners' entries even in the same namespace.

Validation

The server accepts any non-empty string as a namespace. There is no length cap, no character whitelist, no normalization (whitespace, case, Unicode). Whatever you send is stored verbatim and matched with exact equality. If you omit the namespace, the server falls back to the literal string "default".

Implication: "my-app", " my-app" (leading space), "My-App", and "my-app/" are four distinct namespaces. Pick a convention and stick to it.

Flat, not hierarchical

Slashes and dots have no special meaning. "chat/user-42" is a single opaque label, not a path. The server uses WHERE namespace = $1 exact-equality for every read; there is no prefix matching, no parent/child traversal, and no wildcard query. If you need hierarchy, build it in the application layer (e.g. recall across known namespaces and merge client-side).

Overwrite behavior — remember() is always append, never upsert

Every accepted remember() call creates a new memory entry with a freshly generated UUID. Sending the same text to the same (owner, namespace) twice will produce two separate entries that both surface in future recalls. The namespace is metadata for filtering, not a key for deduplication.

await memwal.remember("I prefer dark mode", "prefs");
await memwal.remember("I prefer dark mode", "prefs");
// recall("preferences", { namespace: "prefs" }) → 2 entries, both with the same text

If you need uniqueness, either dedupe before calling remember(), or delete the prior entry first.

Isolation guarantees

Scenario	Visible to recall?
Same owner, same namespace	✅
Same owner, different namespace	❌
Different owner, same namespace	❌
Different owner, different namespace	❌

Cross-namespace and cross-owner reads are not just filtered out of results — the server's SQL WHERE clause excludes them entirely, so they are never decrypted or transferred.

Restore Semantics

restore(namespace, limit?) rebuilds missing local index entries for a namespace from Walrus. It is a recovery operation, not a sync — already-indexed blobs are left alone.

Response fields

Field	Counts	Notes
restored	Blobs the relayer just rebuilt this call	Pulled from Walrus → SEAL decrypted → re-embedded → inserted as a new row
skipped	On-chain blobs already in the local index	No work needed; relayer left them as-is
total	All on-chain blobs the relayer saw for (owner, namespace)	Before the limit was applied
namespace	Echo of the request
owner	Resolved owner address

Silent drops. A blob that cannot be decrypted or embedded (e.g. wrong delegate key, malformed ciphertext, embedding API down) is dropped without counting in restored or skipped. restored + skipped is therefore a lower bound on healthy entries, not a strict equality with total.

Default and limit

• limit defaults to 10 in both TypeScript and Python SDKs and matches the server-side default. The Python SDK historically defaulted to 50; it is now realigned with the server.
• limit caps the inspected blob set, newest-first. It does not cap restored independently — if all 10 inspected blobs are already indexed, restored = 0 and skipped = 10.
• There is no enforced server-side maximum, but very large limits will dominate latency (see below).

Pagination

Restore is single-shot — there is no cursor. To rebuild a namespace larger than your chosen limit, call again with a larger limit, or delete local rows you want re-imported first. Pagination is on the roadmap; until it lands, treat restore() as a "top up to N most recent" operation.

Performance

Latency scales linearly in limit:

• Up to 10 concurrent Walrus aggregator downloads
• Up to 3 concurrent SEAL decrypts (CPU-bound, capped intentionally)
• Embedding requests in parallel (bounded by the relayer's embedding pool)

Expect seconds per blob on a cold cache. Use small limits (≤ 50) for interactive flows and run larger restores out-of-band.

Recall Distance and Filtering

recall() returns the closest K memories by vector distance. There is no default relevance threshold, so small namespaces may return weak filler results because they are still the closest available matches.

Lower distance means more similar:

Distance	Rough meaning
< 0.25	Duplicate or very close
0.25 - 0.55	Related
0.55 - 0.7	Weak/noisy
>= 0.7	Usually unrelated

Use SDK-side filtering when you only want clearly relevant results:

const memories = await memwal.recall({
  query: "what did I eat yesterday?",
  limit: 10,
  namespace: "reading-tracker",
  maxDistance: 0.7,
});

Equivalent manual filtering:

const memories = await memwal.recall({
  query: "what did I eat yesterday?",
  limit: 10,
  namespace: "reading-tracker",
});
const relevant = memories.results.filter((memory) => memory.distance < 0.7);

---

Configuration

MemWalConfig

Field	Type	Required	Default	Description
key	string	Yes	—	Ed25519 delegate private key in hex
accountId	string	Yes	—	Walrus Memory account object ID on Sui
serverUrl	string	No	https://relayer.memwal.ai	Relayer URL
namespace	string	No	"default"	Default namespace for memory isolation

Managed Relayer Endpoints

Network	Relayer URL
Production (mainnet)	https://relayer.memwal.ai
Staging (testnet)	https://relayer.staging.memwal.ai

Framework and Key Handling

Delegate private keys belong on the server only. In Next.js App Router, call Walrus Memory from server actions, route handlers, or other server-only modules that read MEMWAL_PRIVATE_KEY from server env.

"use server" files can only export async functions; keep constants, schemas, and reusable client builders in a separate server-only module.

// app/actions/memory.ts
"use server";

import { getMemWal } from "@/lib/memwal";

export async function savePreference(text: string) {
  const memwal = getMemWal();
  return memwal.rememberAndWait(text, "my-app", { timeoutMs: 30_000 });
}

// lib/memwal.ts
import "server-only";
import { MemWal } from "@mysten-incubation/memwal";

export function getMemWal() {
  return MemWal.create({
    key: process.env.MEMWAL_PRIVATE_KEY!,
    accountId: process.env.MEMWAL_ACCOUNT_ID!,
    serverUrl: process.env.MEMWAL_SERVER_URL ?? "https://relayer.memwal.ai",
    namespace: "my-app",
  });
}

Namespace strategy: owner + namespace is the isolation boundary. Use one namespace per app by default, then split by user, team, or feature when a single app needs separate memory spaces.

Relayer choice: use staging/testnet for learning and prototypes; use production/mainnet for production data. Do not mix staging credentials with mainnet relayer configs.

---

Vercel AI SDK Integration

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:

• Recalls relevant memories before generation
• Extracts and saves facts from conversations after generation

---

OpenClaw / NemoClaw Plugin

For OpenClaw agent integration, use the @mysten-incubation/oc-memwal plugin.

Install

openclaw plugins install @mysten-incubation/oc-memwal

Configure

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 context
• before_reset — saves session summary
• agent_end — captures last response

---

Troubleshooting

Symptom	Fix
health() returns error	Check relayer URL is correct and reachable
recall() returns empty	Verify namespace matches what was used in remember()
recall() returns unrelated filler	Recall is top-K without a default relevance threshold; filter by distance, for example distance < 0.7
401 Unauthorized	Usually wrong MEMWAL_PRIVATE_KEY, key not registered on the account, account ID mismatch, or staging/mainnet mismatch. Check .env.local and dashboard credentials
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
Direct Sui reads fail or examples look stale	Prefer SuiGrpcClient from @mysten/sui/grpc; JSON-RPC snippets using SuiClient / getFullnodeUrl may be stale
forget expectations are unclear	Current relayer POST /api/forget removes vector index rows so memories are unrecallable; Walrus blobs persist until epoch expiry

---

Brand Terminology

Until product confirms a canonical naming pass, these are the working assumptions reflected across this doc, the SDKs, and the relayer. Treat them as descriptive, not authoritative.

Surface	Canonical term	Notes
Product / docs / UI	Walrus Memory	Used in marketing copy, user-facing dashboards, and prose docs
Package / env vars / internal shorthand	memwal	Used in @mysten-incubation/memwal, pip install memwal, MEMWAL_* env vars, internal logs, and codepaths

If you're writing user-facing copy, prefer "Walrus Memory". If you're writing an env var, import path, or grep-target, prefer memwal. Don't mass-rename existing identifiers — that requires a coordinated migration outside this skill's scope.

Links

• Docs: https://docs.memwal.ai
• SDK on npm: https://www.npmjs.com/package/@mysten-incubation/memwal
• GitHub: https://github.com/CommandOSSLabs/MemWal
• Dashboard: https://memwal.ai
• llms.txt: https://docs.memwal.ai/llms.txt