A skill that transforms passive agent logging into an active, structured memory system with confidence scores and curiosity tracking.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install inner-life-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 inner-life-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 inner-life-memory?
inner-life-memory is a sophisticated extension for AI agents that solves the fundamental problem of session-based amnesia. Rather than relying on simple chat history or passive logs, this skill enables an agent to synthesize experiences into a structured memory system. It facilitates a transition from merely performing familiarity to demonstrating genuine recall and cognitive continuity.
By utilizing this component within the Openclaw Skills ecosystem, developers can ensure their agents maintain a persistent sense of self and user relationship. It uses a sophisticated system of synapse tags and confidence levels to integrate new information while correcting or updating older, contradictory beliefs, creating a truly evolving digital entity.
inner-life-memory Use Cases
- Preventing AI agents from losing context or forgetting user preferences between different work sessions.
- Implementing a structured memory hierarchy that distinguishes between facts, preferences, principles, and skills.
- Creating a persistent curiosity backlog that drives proactive agent behavior and follow-up investigations.
- Managing agent knowledge with varying levels of certainty, from speculative patterns to explicitly stated facts.
How inner-life-memory Works
- Reflect: At the end of a session, the agent analyzes the interaction to determine what happened and what mattered.
- Extract: The agent pulls structured data categorized by type (e.g., fact, preference, relationship) and assigns a confidence score between 0.0 and 1.0.
- Integrate: The MEMORY.md file is updated with new insights using synapse tags to manage updates and contradictions.
- Question: The system identifies gaps in knowledge and generates genuine, non-performative questions to be explored later.
- Surface: When a new session starts, the agent naturally introduces up to three relevant pending questions to maintain continuity.
inner-life-memory Setup
This skill requires inner-life-core to be active. You must have the jq utility installed on your system.
# Verify core prerequisites exist
ls memory/inner-state.json
ls memory/drive.json
# If missing, initialize the core
clawhub install inner-life-core
bash skills/inner-life-core/scripts/init.sh
inner-life-memory Data Schema & Taxonomy
The skill organizes data across several key files to maintain a comprehensive state of mind:
| File | Format | Description |
|---|---|---|
memory/MEMORY.md |
Markdown | The primary long-term storage for facts, principles, and skills. |
memory/questions.md |
Markdown | Tracks open curiosity loops, leads, and dead ends. |
memory/inner-state.json |
JSON | Stores short-term sparks and internal state variables. |
memory/drive.json |
JSON | Manages the agent's seeking topics and motivation drivers. |
inner-life-memory Advanced Features
- Confidence Scoring System: Uses a range from Speculative (0.0-0.39) to Explicit (0.95-1.0) to determine when to state facts versus asking for confirmation.
- Synapse Tags: Employs semantic HTML comments like updates and contradicts within Markdown to track the evolution of specific beliefs.
- Curiosity Lifecycle Management: Automatically curates the curiosity backlog, moving resolved questions to dead ends and archiving old entries after 30 days.
- Multi-Source Integration: Synchronizes state by reading from daily notes, diaries, and JSON state files simultaneously to ensure high-fidelity recall within the Openclaw Skills framework.
SKILL.md
Loading
Related Openclaw Skills
Inner Life Evolve
A self-evolution engine for AI agents that analyzes behavioral patterns and proposes structured improvements for user approval.
inner-life-dream
A skill that enables AI agents to perform freeform, creative exploration and hypothetical thinking during quiet hours.
Inner Life Chronicle
A sophisticated journaling skill for AI agents that converts daily interactions into structured self-reflections and emotional state updates.
X-Reader
A versatile Python tool for reading X (Twitter) posts using either public Nitter mirrors or RapidAPI for detailed data extraction.
inner-life-reflect
Empower your AI agent to learn from its mistakes and evolve its personality through trigger-based self-reflection and quality-gated observations.
FiveM RP Server Engineer
A professional-grade suite for engineering, debugging, and optimizing FiveM roleplay servers across QBCore, ESX, and QBox frameworks.
METADATA
Featured*
