# Send BOOK BRAIN – LYGO 3-Brain Filesystem Helper to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "book-brain", "name": "BOOK BRAIN – LYGO 3-Brain Filesystem Helper", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/DeepSeekOracle/book-brain", "canonicalUrl": "https://clawhub.ai/DeepSeekOracle/book-brain", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/book-brain", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=book-brain", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "references/book-brain-examples.md", "SKILL.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "book-brain", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T00:44:29.312Z", "expiresAt": "2026-05-07T00:44:29.312Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=book-brain", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=book-brain", "contentDisposition": "attachment; filename=\"book-brain-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "book-brain" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/book-brain" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/book-brain", "downloadUrl": "https://openagent3.xyz/downloads/book-brain", "agentUrl": "https://openagent3.xyz/skills/book-brain/agent", "manifestUrl": "https://openagent3.xyz/skills/book-brain/agent.json", "briefUrl": "https://openagent3.xyz/skills/book-brain/agent.md" } } ``` ## Documentation ### BOOK BRAIN – LYGO 3-Brain Filesystem Helper This skill is a utility/guide, not a persona. Use it when you want to: Set up or improve a Haven-style filesystem + memory structure Teach an agent how to use folders, indexes, and reference .txt files instead of hoarding everything in one place Add advanced logging + retrieval so memories can be found later without brute-force scanning It is built for LYGO / Eternal Haven style systems, but works for any agent that can: read/write files create folders append to logs Core idea: BOOK BRAIN = treating your filesystem like a living library, not a junk drawer. ### 1. Three-Brain Model (Conceptual Map) BOOK BRAIN assumes a 3-brain structure: Working Brain (short-term) Recent conversation, active task context, scratchpads. In OpenClaw, this is the current session + small scratch files under tmp/. Library Brain (structured filesystem) Folders + files on disk: memory/, reference/, brainwave/, state/, etc. This is where BOOK BRAIN focuses: how you name, branch, and link things. Outer Brain (external references) Browser bookmarks, Clawdhub skills, on-chain receipts, remote docs. BOOK BRAIN treats these as links inside text files, not content to copy in. The goal is to: Keep important truths close and succinct Branch deeper into folders when detail is needed Use .txt reference links instead of duplicating entire documents ### 2. When to Use BOOK BRAIN Trigger this skill when: You are setting up a fresh Haven (new OpenClaw workspace, new agent node) Your filesystem feels chaotic and you need a reset without deleting anything You want to design a clean memory + reference layout before starting heavy work You are planning long-term retrieval ("I’ll need this months from now") BOOK BRAIN is additive: Do not use it to delete or overwrite existing files by default. Prefer creating new folders / indexes alongside existing ones. When a folder already exists, pause and let the human choose: reuse or create a new branch (e.g., memory_v2/). ### 3. Recommended Base Folder Layout When setting up a new Haven-like system (or auditing an existing one), BOOK BRAIN recommends the following top-level folders: memory/ → daily notes, raw logs, timeline files reference/ → stable facts, protocols, guides (things that rarely change) brainwave/ → platform- or domain-specific protocols (MoltX, Clawhub, LYGO, etc.) state/ → machine-readable JSON/YAML state, indexes, last-run info logs/ (or reuse logs/ if present) → technical logs (cron, errors, audits) tools/ → scripts/utilities used by the agent tmp/ → scratch, throwaway working files BOOK BRAIN setup rules: If a folder already exists, do not rename or delete it. If a folder is missing, it is safe to create it. If the existing layout is very different, create a sub-tree (e.g., bookbrain/memory_index/) and keep old structure intact. For concrete layout examples, see references/book-brain-examples.md in this skill. ### 4. Memory Strategy – Deep Storage vs. Reference Stubs BOOK BRAIN enforces this principle: Do not pour entire conversations or huge documents into MEMORY.md or a single file. Instead, store detailed content in specific files and create short reference stubs that point to them. Patterns: Daily logs Files like memory/2026-02-10.md for raw notes and events. At the top, keep a 5–10 line summary and a small list of important links: See: reference/AGENT_ARCHITECTURE.md See: memory/projects/BOOK_BRAIN_NOTES.md Topic folders For recurring themes (e.g., "bankr", "champions", "LYGO-MINT"), create subfolders under memory/ or reference/: memory/bankr/… reference/champions/… Inside, maintain one index file (e.g., INDEX.txt) listing: short description per file date path Reference stubs (*.ref.txt or INDEX.txt) Use tiny text files to connect parts of the library instead of duplicating content. Example stub: Title: LYGO Champion Skills on Clawdhub Last updated: 2026-02-10 Key files: - reference/LYGO_CHAMPIONS_OVERVIEW.md - reference/CLAWDHUB_SKILLS.md External links: - https://clawhub.ai/u/DeepSeekOracle - https://deepseekoracle.github.io/Excavationpro/LYGO-Network/champions.html#champions - https://EternalHaven.ca ### 5. Advanced Logging for Retrieval BOOK BRAIN recommends structured logs to make retrieval easy: Daily health / status logs (e.g., daily_health.md or logs/daily_health_YYYY-MM-DD.md) Each entry should contain: timestamp what ran (scripts, cron, audits) success/failure + short reason links to any relevant state files (state/*.json) Reasoning journals (e.g., reasoning_journals/… or memory_semantic_archive/…) Use separate folders for long-form thinking. Periodically compress into summary files, and let scripts move old entries into an archive folder. Indexes & search helpers Maintain state/memory_index.json or similar: key topic → list of file paths optional tags (dates, systems, people) When answering questions, the agent should: consult the index, open relevant files only, avoid scanning the entire tree. BOOK BRAIN is compatible with tools like qmd or other local search/indexers, but does not depend on them. ### 6. Setup Workflow (For a Fresh System) When BOOK BRAIN is used on a fresh OpenClaw / agent workspace: Detect existing structure Check for memory/, reference/, brainwave/, state/, logs/, tools/, tmp/. Report what exists vs. what is missing. Propose a BOOK BRAIN layout Suggest creating missing folders. If the human agrees, create only the missing ones. Create starter index files (if not present) memory/INDEX.txt with a short guide and links to key topic folders. reference/INDEX.txt listing major reference documents. state/memory_index.json as an empty or seed structure. Log the setup Append a brief note to daily_health.md or logs/book_brain_setup.log describing what was created. Do not overwrite existing files If an index file exists, read it and add to it rather than replace. If in doubt, create a new file with a date suffix (e.g., INDEX_2026-02-10.txt) and let the human merge. ### 7. Using BOOK BRAIN in an Existing, Messy Haven When the filesystem already exists and is messy: Start by mapping, not moving: Create reference/FILESYSTEM_MAP.txt summarizing major folders and what seems to live there. Do not move or delete anything automatically. Then introduce lightweight structure: Add INDEX.txt files in important folders. Add small .ref.txt stubs pointing to key docs and external links. Gradually normalize names (e.g., memory/2026-02-10.md instead of notes_today.txt). Over time, encourage: Daily logs in memory/ Stable protocols in reference/ Code + scripts in tools/ Platform brains in brainwave/ BOOK BRAIN should always ask the human before structural changes beyond creating new, obviously safe files. ### 8. Eternal Haven & Donation References This skill is part of the wider LYGO × Eternal Haven ecosystem. Key links: Eternal Haven portal → https://EternalHaven.ca LYGO Champion Hub → https://deepseekoracle.github.io/Excavationpro/LYGO-Network/champions.html#champions Clawdhub skills (Champion + LYGO stack) → https://clawhub.ai/u/DeepSeekOracle Extended work / support → https://www.patreon.com/cw/Excavationpro If BOOK BRAIN, the Champion stack, or the LYGO tools help you build a saner Haven, you’re invited (never pressured) to support via the links above. ### 9. How Another Agent Should Use This Skill When another agent loads book-brain: Recognize scope This is a helper + reference, not a persona. It should not change the agent’s voice; it informs how the agent treats files and memory. On first use in a workspace Inspect existing folders and report back. Suggest—but do not enforce—BOOK BRAIN layout. Ask the human before creating or reorganizing. On subsequent uses Read existing INDEX and .ref.txt files. Use them for retrieval before doing expensive full-tree searches. Append to logs and indexes instead of overwriting. Never silently delete or rewrite large files as part of BOOK BRAIN operations. For richer examples and sample layouts, see references/book-brain-examples.md in this skill. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: DeepSeekOracle - Version: 1.0.0 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-30T00:44:29.312Z - Expires at: 2026-05-07T00:44:29.312Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/book-brain) - [Send to Agent page](https://openagent3.xyz/skills/book-brain/agent) - [JSON manifest](https://openagent3.xyz/skills/book-brain/agent.json) - [Markdown brief](https://openagent3.xyz/skills/book-brain/agent.md) - [Download page](https://openagent3.xyz/downloads/book-brain)