nox-auto-capture

Capture user corrections, decisions, facts, and lessons into a Qdrant vector database.

3|1|Updated Mar 8, 2026

Search Tags:#automation #learning #memory #conversation #qdrant #capture

Ready to run

npx skills add https://github.com/rockywuest/openclaw-memory-local --skill nox-auto-capture

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nox-auto-capture
Source: https://github.com/rockywuest/openclaw-memory-local/tree/main/plugins/nox-auto-capture
Command: npx skills add https://github.com/rockywuest/openclaw-memory-local --skill nox-auto-capture

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of capturing valuable information from agent conversations, eliminating the need for manual memory management and ensuring the agent continuously learns from interactions.

Core Features & Use Cases

Automatic Capture: Silently listens for and captures corrections, decisions, new facts, and lessons learned from user messages.
Intelligent Filtering: Employs over 30 skip patterns to filter out system noise, brief messages, and irrelevant content.
Deduplication: Uses SHA256 hashing to prevent duplicate information from being stored.
Use Case: As you interact with your agent, it will automatically save key decisions you make, facts you state, and corrections you provide, building a rich, context-aware memory without you having to explicitly tell it to remember.

Quick Start

Configure the nox-auto-capture plugin in your openclaw.json file to enable automatic memory capture.

Frequently Asked Questions about nox-auto-capture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I make an agent learn from conversations automatically without manual memory management?▼

Automatic conversation capture enables agents to learn from interactions by silently extracting corrections, decisions, and facts from user messages into a local Qdrant vector database. This eliminates manual memory management by filtering system noise with over 30 skip patterns.

What is the best way to capture lessons learned from agent conversations into a vector database?▼

Capturing lessons learned from agent conversations is best handled by automatically extracting user-identified facts and storing them in a Qdrant vector database. The process uses SHA256 hashing for deduplication and intelligent filtering to ensure only valuable information is saved.

Does automatic conversation capture filter out system noise and brief messages?▼

Automatic conversation capture filters system noise and brief messages using over 30 skip patterns. It exclusively captures user messages, ensuring that only meaningful corrections, decisions, and facts are stored in the local vector database.

How do I prevent duplicate information from being stored when capturing agent conversation memory?▼

Preventing duplicate information when capturing agent conversation memory is achieved through SHA256 hashing. This deduplication mechanism ensures that identical facts, corrections, or decisions are not repeatedly stored in the Qdrant vector database.

Can I configure automatic memory capture for agents using a local Qdrant database?▼

You can configure automatic memory capture for agents using a local Qdrant database by setting up the capture plugin in your openclaw.json file. This enables the agent to silently save key decisions and facts during interactions.

When should I not use automatic conversation capture for agent memory?▼

Automatic conversation capture should not be used when you need explicit, manual control over what the agent remembers. Because it only captures user messages and relies on pattern filtering, it may skip context that does not match its predefined extraction logic.