A cognitive architecture for AI agents that synchronizes visual perception with a structured 3-brain memory system.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install book-brain-visual-reader
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 book-brain-visual-reader 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 Book Brain Visual Reader?
Book Brain Visual Reader is an advanced cognitive utility designed for AI agents that possess visual capabilities, such as browser automation, OCR, or image analysis tools. It expands the traditional 3-brain memory model by introducing a dual-hemisphere reasoning protocol. This allows agents to process structured data through the Left Brain while simultaneously analyzing visual contexts like screenshots and UI layouts through the Right Brain.
By implementing this framework from Openclaw Skills, developers can ensure their agents maintain a non-destructive filesystem while performing high-fidelity verification between what an API reports and what a browser actually displays. This skill is essential for building reliable Openclaw Skills that require visual evidence gathering, such as auditing dashboards, capturing on-chain receipts, or managing complex digital workspaces.
Book Brain Visual Reader Use Cases
- Verifying live web dashboard data against backend API responses to detect UI discrepancies.
- Organizing visual evidence like screenshots, seals, and diagrams within a permanent Library Brain.
- Implementing a structured memory system for browser-based agents to prevent data loss.
- Cross-referencing technical logs with visual snapshots for comprehensive system auditing.
How Book Brain Visual Reader Works
- The agent initializes a 3-brain filesystem including Working, Library, and Outer brain segments with a dedicated visual directory.
- The Left Brain phase executes, retrieving structured text, JSON schemas, or API data as the baseline expectation.
- The Right Brain phase triggers, capturing visual snapshots and analyzing spatial patterns or visual anomalies.
- The agent reconciles both hemispheres, comparing the textual data against visual evidence to confirm accuracy.
- The final verified data is logged into the Library Brain using reference stubs and visual indexes for future retrieval.
Book Brain Visual Reader Setup
To deploy this memory structure for your Openclaw Skills, execute the following commands in your agent's workspace:
# Create the visual-aware filesystem hierarchy
mkdir -p memory reference brainwave state logs tools tmp visual/screenshots visual/dashboards
# Initialize the visual tracking index
touch reference/VISUAL_INDEX.txt
# Log the framework activation
echo "Book Brain Visual Reader initialized" >> logs/setup.log
Book Brain Visual Reader Data Schema & Taxonomy
The Book Brain Visual Reader organizes data using a hierarchical taxonomy to ensure no information is overwritten:
| Directory | Data Type | Description |
|---|---|---|
visual/ |
Binary/Images | Long-term storage for screenshots, dashboards, and visual seals. |
reference/ |
Markdown/Text | Stable protocols, visual indexes, and external URL stubs. |
state/ |
JSON | Machine-readable metadata, hashes, and memory indexes. |
memory/ |
Logs/Notes | Daily journals and raw activity captures. |
brainwave/ |
Protocols | Domain-specific rules for platforms like LYGO or MoltX. |
Book Brain Visual Reader Advanced Features
- Dual-hemisphere '5D' reasoning (Text + Visual + API + State + Timeline) for maximum data integrity.
- Non-destructive filesystem architecture ensuring all data captures are additive rather than subtractive.
- Reference stubs for the Outer Brain that link external web resources without bloating the agent context.
- Automated visual indexing to map screenshots to specific topics or receipt timelines.
- Support for multi-agent collaboration within the same Openclaw Skills memory environment.
SKILL.md
Loading
Related Openclaw Skills
Sales Rhythm Tracker
A B2B sales pipeline manager for Openclaw agents based on the legendary Alibaba Iron Army methodology.
Whale Alert
A professional on-chain monitoring tool for tracking high-value cryptocurrency transfers and identifying smart money movements.
Wallet Analyzer
A comprehensive cryptocurrency wallet analysis tool for tracking holdings, P&L, and suspicious activities across major blockchains.
Blockchain Transaction Decoder
A professional-grade tool for decoding raw blockchain transactions into human-readable contract interactions and financial records.
LYGO-MINT Operator Suite
A technical toolkit for canonicalizing, hashing, and anchoring multi-file AI agent packs into verifiable, receipts-first ledgers.
OpenClaw Flow Kit
A comprehensive utility suite designed to resolve workflow bottlenecks, handle rate limits, and standardize script execution for AI agents.
Featured*
