Viking Memory

ReviewAudited by ClawScan on May 10, 2026.

Overview

This is a coherent local memory skill, but it can automatically save and later reuse potentially sensitive long-term memories without clear approval, scope, or retention controls.

Install only if you want OpenClaw to use a local OpenViking long-term memory store. Configure the agent to ask before saving memories, avoid storing secrets or sensitive personal details, and make sure you know how to review and delete saved memories.

Findings (2)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Private or sensitive details from conversations could be stored and later reused by the agent without the user realizing exactly what was saved.

Why it was flagged

The skill explicitly handles user preferences, conversation history, and important information as long-term memory, including automatic saving, but does not document user confirmation, retention, deletion, or reuse boundaries.

Skill content
用于语义检索用户偏好、历史对话、重要信息... 对话中识别到重要信息 → 自动保存
Recommendation

Require explicit user confirmation before saving memories, clearly label what will be stored, document retention/deletion controls, and avoid storing secrets or sensitive personal data.

What this means

The agent may add or change what it remembers about the user, which can affect future responses.

Why it was flagged

The code exposes a write operation to the local memory API. This is purpose-aligned for a memory skill, but it gives the agent authority to mutate persistent memory contents.

Skill content
const result = await vikingRequest('/api/v1/resources', { uri, content, });
Recommendation

Use this skill with a workflow that asks before writing important memories and allows users to inspect or remove saved entries.