A specialized procedural memory skill that enables AI agents to form habits and preferences through repetitive actions and reward-based learning.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install basal-ganglia-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 basal-ganglia-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 Basal Ganglia Memory?
Basal Ganglia Memory is a core component of the AI Brain series, specifically engineered to simulate the biological basal ganglia's role in habit formation and procedural memory. This skill allows AI agents to evolve beyond simple prompt-response cycles by developing internal preferences and "muscle memory" for frequently executed workflows. By integrating this into Openclaw Skills, developers can create more efficient agents that recognize and automate patterns that have historically led to successful outcomes.
As part of a modular neural architecture, this skill works alongside other cognitive modules to provide a comprehensive memory layer. It transforms one-off actions into stable procedural habits, significantly reducing the cognitive load on LLMs during routine tasks and establishing a persistent personality through consistent behavioral preferences.
Basal Ganglia Memory Use Cases
- Automating recurring developer workflows by identifying repeated command patterns.
- Establishing user-specific preferences for code formatting or architectural choices.
- Reinforcing successful task completion paths through reward-based feedback loops.
- Reducing latency in complex operations by utilizing procedural shortcuts for common triggers.
How Basal Ganglia Memory Works
- The skill monitors the agent's actions to identify patterns and sequences that occur frequently.
- Successful outcomes generate positive reinforcement, increasing the "reward score" for specific procedural paths.
- Frequently reinforced paths are transitioned into a habit state, where they become the default preference for the agent.
- When a familiar scenario is triggered, the agent prioritizes its established procedural memory over generating a new logic path from scratch.
- The skill continuously refines these habits based on ongoing interaction and success rates.
Basal Ganglia Memory Setup
This skill is currently under development as part of the AI Brain series. To monitor progress or participate in early testing within the Openclaw Skills ecosystem, follow these steps:
# Clone the repository
git clone https://github.com/ImpKind/basal-ganglia-memory.git
# Navigate to the skill directory
cd basal-ganglia-memory
# Install dependencies
npm install
Basal Ganglia Memory Data Schema & Taxonomy
The skill organizes procedural data to optimize for retrieval and reinforcement. The following schema represents the primary habit storage structure:
| Attribute | Type | Description |
|---|---|---|
habit_id |
String | Unique identifier for a specific procedural sequence. |
activation_count |
Integer | Number of times this habit has been triggered. |
reward_score |
Float | The current weight of the habit based on reinforcement. |
procedure_data |
JSON | The specific steps, commands, or logic comprising the habit. |
last_used |
Timestamp | ISO 8601 string of the most recent activation. |
Basal Ganglia Memory Advanced Features
- Multi-agent synchronization to share procedural habits across a fleet of Openclaw Skills agents.
- Integration with the Amygdala-Memory skill to weight habits based on emotional or high-priority context.
- Custom reward function hooks to manually influence how agents prioritize specific procedural paths.
- Decay algorithms that allow outdated or unsuccessful habits to eventually be pruned from the procedural memory.
SKILL.md
Loading
Related Openclaw Skills
Anterior Cingulate Memory
A neural-inspired conflict detection and error monitoring module that acts as a cognitive filter for AI agents.
Anterior Cingulate Memory
A conflict detection and error monitoring system that enables AI agents to learn from past mistakes and avoid recurring patterns.
VPS Setup for Production
A comprehensive automation skill for deploying a production-ready VPS environment featuring Docker Swarm, Traefik v3, and Portainer CE.
Evolution API v2.3 WhatsApp Automation
A robust WhatsApp automation engine for managing instances, sending interactive messages, and integrating AI chatbots.
Insula Memory
A neuro-inspired internal state awareness module for AI agents to track energy, mood, and interoception.
AgentMesh Governance
A zero-trust governance layer providing cryptographic identity verification, policy enforcement, and tamper-evident auditing for AI agents.
Featured*
