A file-based knowledge organization and retrieval system designed to capture URLs, bookmarks, summaries, and agent research outputs instantly.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install super-agent-knowledge
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 super-agent-knowledge 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 Super Agent Knowledge Capture?
Super Agent Knowledge Capture is a lightweight, local-first knowledge management system designed specifically for AI agents and developers. This utility allows agents to immediately capture, index, and query external data including web URLs, video and article extracts, social media posts, and custom agent-generated research outputs. By storing assets in clean Markdown files with standardized YAML frontmatter, it bridges the gap between raw web discoveries and structured long-term agent memory.
As part of the broader ecosystem of Openclaw Skills, this tool ensures your AI workflows remain tidy and accessible. Rather than losing transient insights in fleeting session logs, agents can use this skill to programmatically append data to structured local directories and maintain an automated, searchable index file for seamless future retrieval.
Super Agent Knowledge Capture Use Cases
- Preserving web articles, videos, and academic papers during deep agent-driven analysis.
- Saving and tag-indexing social media posts and threads for future context synthesis.
- Storing and indexing agent-generated research reports directly in a local workspace.
- Preventing valuable agent context from being lost in daily ephemeral memory logs.
- Building a local, searchable secondary brain for autonomous workflows.
How Super Agent Knowledge Capture Works
- Capture Command: The CLI tool is executed by the user or agent to capture a resource (e.g.,
know add url). - Frontmatter Generation: The skill automatically structures metadata into a standardized YAML frontmatter format (containing type, title, source, tags, and summary).
- Directory Routing: Files are routed to specific subdirectories (
urls/,extracts/,posts/, orresearch/) based on content type. - Auto-indexing: The master
INDEX.mdfile is instantly updated to provide a human- and agent-readable index of all entries. - Maintenance and Normalization: Automated cleanup tools normalize tags, validate YAML schemas, and restructure misplaced files to maintain perfect data integrity.
Super Agent Knowledge Capture Setup
Installation
To install Super Agent Knowledge Capture, run the standard CLI command:
clawhub install knowledge
This installs scripts/know. You can add it to your PATH or reference it directly via its installation path:
~/.openclaw/skills/knowledge/scripts/know
Configuration
By default, all knowledge assets are saved to:
~/.soulshare/agent/knowledge/
You can customize this workspace destination by modifying your config file at ~/.config/know/config or by exporting the KNOWLEDGE_DIR environment variable.
Super Agent Knowledge Capture Data Schema & Taxonomy
The workspace maintains a predictable, flat-file structure within your storage directory:
INDEX.md: Auto-maintained directory indexing all saved articles, URLs, and research notes.urls/: Stored bookmark files for captured web links.extracts/: Extracted highlights and summaries from videos, articles, and research papers.posts/: Logged social threads and micro-blogging posts.research/: Structured research notes and synthesis outputs generated by agents.
Metadata Schema (YAML Frontmatter)
All captured files are saved as Markdown containing structured frontmatter to support search indexing and semantic matching:
---
type: url|extract|post|research
title: "Entry title"
source_url: "https://..."
source_kind: url|video|article|paper|post|research
tags: ["tag1", "tag2"]
added: "2026-02-26"
added_by: "agent-name"
summary: "One-line summary"
---
Super Agent Knowledge Capture Advanced Features
- Automated Maintenance (
know tidy): Programmatically audit your data collection, normalize tags, move misplaced files, and purge empty entries in a single step. - QMD Integration: Turn plain markdown directories into QMD-indexable vector and text spaces for semantic search queries via agent protocols.
- Flexible Tag Filtering: Run granular queries natively using the CLI (e.g.,
know list --tags security) to retrieve specific slices of your knowledge base. - Workflow Integration: Schedule
know tidy --fixto run inside developer heartbeat loops or nightly cron jobs to ensure continuous database health.
SKILL.md
Loading
Related Openclaw Skills
Super Claw Local Knowledge
A local knowledge base skill enabling AI agents to seamlessly ingest, parse, and retrieve contextual information from corporate and personal multi-format documents.
super-Finance Knowledge Base
An AI-driven financial knowledge management system featuring dynamic organization, semantic search, and RAG capabilities for institutional insights.
super-Knowledge Capture
Seamlessly transform raw team conversations and transcripts into structured, searchable Notion pages.
super-Taizi Knowledge Base
A unified personal knowledge management skill that seamlessly integrates vector search, entity relationship mapping, and Obsidian-style note storage.
Kano Model Strategist & Feature Pruner
An AI-powered product strategy tool that classifies features into Kano categories to cut development waste and prioritize Minimum Delightful Products (MDP).
yescan-universal
An advanced CLI-driven image processing, OCR, and document conversion skill powered by Quark Scan King.
Featured*
