More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    optimal-brain vault pack

    1

    Turn an Obsidian vault into durable long-term memory for research, repo seeding, and sourced technical writing.

    Free

    1 useSecurity scanned
    optimal-brain vault pack

    optimal-brain vault pack

    Example session with this skill installed

    I need to research our current implementation of the auth flow from my notes and then draft a technical memo on how we can improve it. Use my Obsidian vault at /Users/dev/Documents/Knowledge.

    • Read your context and instructions
    • Compiled the optimal-brain vault

    I have initiated the research session and updated your knowledge vault. I scanned the Auth and Security folders in your vault, identified three existing implementation notes, and synthesized the findings into a new research hub.

    Research & Draft Summary

    TaskStatusLocation
    Research SessionCreatedResearch Sessions/2023-11-24 Auth Flow Review.md
    Findings Captured4 NotesFindings/Auth-JWT-Logic.md, Findings/Auth-Middleware-Gap.md
    Claim LedgerGeneratedDrafts/Auth Improvement Memo/Claim Ledger.md
    Draft MemoStage 1Drafts/Auth Improvement Memo/Brief.md

    Research Synthesis: Auth Flow Improvements

    Based on your notes in [[Project-X Auth Spec]] and [[Security Audit 2023]], the current implementation relies on a single-factor JWT flow that lacks refresh token rotation. The vault notes indicate this was a "temporary measure" from six months ago. I have drafted the technical memo brief focusing on the transition to OIDC.

    Next steps

    1. Review the Claim Ledger in your vault to verify the evidence sourced from your personal notes.
    2. Run write-from-vault stage 2 to convert the current fragments into a structured article outline.
    3. Open the

    Context Graph.canvas in Obsidian to see how these new auth findings link to your existing infrastructure notes.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The Optimal Brain Vault Pack solves the problem of transient agent memory. Most coding agents lose context once a session ends or a context window fills up. This skill pack bridges that gap by turning a local Obsidian vault into a durable, structured long-term memory. It allows your agent to research deeply, cite specific sources from your notes and PDFs, and maintain a persistent knowledge graph across different coding sessions.

    What it does

    • Vault orchestration routes between research, seeding, writing, and visualization tasks.
    • Source-backed research extracts and synthesizes data from PDFs and personal notes with full citations.
    • Codebase seeding transforms repo structures, ADRs, and documentation into linked vault notes.
    • Article generation converts raw research fragments into sourced, publishable articles using a claim ledger.
    • Visual context mapping generates Obsidian JSON Canvases to show how project concepts connect.
    • Personal note capture saves your reflections and todos without agent interference.

    How it works

    1. Run the setup command to link your absolute vault path and define project conventions.
    2. Seed the vault from your current repository to give the agent a baseline of your codebase.
    3. Invoke research or writing tasks that pull from and write back to your Obsidian notes.
    4. Maintain the context graph to visualize the relationship between new findings and existing knowledge.

    Frameworks & tools

    This tool is built for Obsidian and any local Markdown-based knowledge base. It utilizes git for repo seeding, pdftotext for document extraction, and the JSON Canvas spec for visual mapping.

    Why this beats prompting it yourself

    Standard prompts lack the "discipline" required for durable memory, often hallucinating citations or failing to follow complex folder hierarchies. This skill enforces a strict capture format, claim ledgers, and validity rules (like superseded_by tags) that keep your long-term knowledge accurate and usable.

    Use cases

    • Onboarding an agent to a massive monorepo by seeding its architecture into a vault.
    • Conducting deep technical research across a library of PDFs and internal documentation.
    • Building a sourced technical blog post or internal memo directly from project notes.
    • Visualizing complex dependency or logic flows using the Obsidian Canvas generator.

    Known limitations

    Does not support cloud-only vaults or web-based note platforms. Requires manual installation of PDF extraction tools for document processing. Agent cannot scan disk for vaults; path must be provided.

    Requires absolute local path for vault. No support for cloud-only notes. PDF extraction requires external CLI tools. Agent only reads/writes configured repo and vault paths.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell

    File Scopes

    optimal-brain-vault-pack/**
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This pack solves the issue of 'agent amnesia' by creating a persistent, structured memory bank in Obsidian, allowing your agent to reference past research and project architecture across entirely different sessions.

    The pack is compatible with any AI agent that supports local file system access and Markdown editing, including popular IDE-based agents and standalone research assistants.

    You will receive the core skill integration files, a set of Obsidian templates optimized for agent 'discipline,' and a configuration guide for linking your local repository and PDF libraries.

    Yes, it requires a local Obsidian installation and the `pdftotext` utility for document extraction; please note that cloud-only platforms like Notion or Roam are not supported.

    Updates are delivered through the marketplace dashboard; because the system uses a standardized Markdown and JSON Canvas structure, your existing notes will remain compatible even as the skill logic evolves.

    Absolutely. The vault orchestration is designed to work with your personal notes and includes a 'Personal Note Capture' feature that ensures the agent does not overwrite or interfere with your private reflections.

    New in Agents & Orchestration