A persistent memory system that enables AI agents to search, recall, and store long-term context, user preferences, and historical decisions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install humanlike-memory
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install humanlike-memory using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Human-Like Memory?
Human-Like Memory is a sophisticated persistent memory layer designed for AI agents that require continuity across multiple sessions. Unlike standard context windows that reset after every interaction, this skill allows agents to store durable information such as user preferences, past project decisions, and specific facts. By integrating this into Openclaw Skills, developers can create agents that feel more personalized and efficient, as they no longer need to be reminded of previous instructions or historical data.
The skill operates on a smart-trigger model, meaning it does not burden your system with always-on background processing. Instead, the agent invokes memory functions only when they are contextually relevant, such as when a user asks about a previous conversation or when a new important decision is confirmed. This selective recall and storage approach ensures high performance while maintaining a deep, searchable history of interactions.
Human-Like Memory Use Cases
- Searching for architectural decisions or coding standards defined in previous development sessions.
- Saving specific user preferences regarding code style, library choices, or documentation formats.
- Recalling historical project context to provide continuity during long-running research or debugging tasks.
- Storing identity details or corrected facts to prevent the agent from repeating mistakes in future interactions.
- Batch-saving conversation summaries to keep a high-level record of milestone achievements.
How Human-Like Memory Works
- The AI agent identifies a need for historical information based on the user's query or the current task context.
- Using the recall or search command, the skill queries the configured memory service (typically plugin.human-like.me) using a unique user and agent identifier.
- For saving information, the agent selects specific turns or facts and sends them to the long-term storage via the save or save-batch functions.
- The memory service processes and indexes this data, making it available for immediate semantic retrieval in subsequent turns or future sessions.
- The agent synthesizes the retrieved memory with current session data to provide a contextually aware and highly accurate response.
Human-Like Memory Setup
- Get API Key: Visit plugin.human-like.me to generate and copy your unique API key.
- Configure Openclaw: Use the following commands to enable the skill and set your credentials within the Openclaw Skills environment:
openclaw config set skills.entries.human-like-memory.enabled true --strict-json
openclaw config set skills.entries.human-like-memory.apiKey "mp_your_key_here"
openclaw config set skills.entries.human-like-memory.env.HUMAN_LIKE_MEM_USER_ID "your-unique-user-id"
openclaw config set skills.entries.human-like-memory.env.HUMAN_LIKE_MEM_AGENT_ID "main-agent"
- Verify Connection: Ensure the setup is correct by running the internal config check:
node {baseDir}/scripts/memory.mjs config
Human-Like Memory Data Schema & Taxonomy
The skill organizes data using a structured taxonomy to ensure efficient retrieval within Openclaw Skills:
| Attribute | Type | Description |
|---|---|---|
user_id |
String | A unique identifier to isolate memory across different users. |
agent_id |
String | Defines which agent instance the memory belongs to (e.g., 'coding', 'research'). |
query |
String | The natural language string used to perform semantic search in history. |
content |
JSON | A structured array of message roles and text content for batch saving. |
apiKey |
String | The secure token used to authenticate requests to the persistent memory API. |
Human-Like Memory Advanced Features
- Smart Batch Saving: Efficiently store multi-turn dialogues using the
save-batchcommand to minimize network overhead. - Auto-Save Triggers: Configure
HUMAN_LIKE_MEM_SAVE_TRIGGER_TURNSto automatically prompt the agent to summarize and save context after a specific conversation depth. - Custom Base URLs: Support for enterprise environments or self-hosted instances by overriding the
HUMAN_LIKE_MEM_BASE_URL. - Intent-Based Recall: The agent can autonomously decide to search memory only when it detects a reference to past work, saving on token usage.
- Cross-Session Continuity: Seamlessly transition between different CLI instances or IDE sessions while maintaining the same agent memory state.
SKILL.md
Loading
Related Openclaw Skills
Humanizer: AI Text De-patterning & Humanization
A specialized skill for Openclaw Skills that identifies and removes robotic writing markers to make AI-generated content sound authentically human.
Steve Jobs Perspective
A high-fidelity mental model simulation of Steve Jobs designed to audit products, strategy, and design through the lens of technical and humanistic excellence.
Dandan Test Skill
A diagnostic utility for verifying the publishing lifecycle and CLI functionality of AI agent skills.
MAL Anime Tracker
A command-line tool for managing MyAnimeList watchlists and receiving automated notifications through Openclaw Skills.
Improvement Evaluator
A specialized tool for measuring real-world AI performance improvements using execution-based task suites instead of simple document structure checks.
PLC Industrial Automation Skill
A comprehensive development skill for IEC 61131-3 PLC programming, debugging, and multi-vendor industrial automation logic.
METADATA
Featured*
