Session Memory Triage: make your agent remember what matters

    1

    This is the method behind Fuli (https://github.com/JevonsCode/fuli), an open-source local memory layer that does this automatically across tools.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Session Memory Triage: make your agent remember what matters

    Example session with this skill installed

    We're done with the deployment script. Remember that I prefer using pnpm over npm, and for this repo, we always deploy to the staging branch first.

    • Read your context and instructions
    • Compiled the session memory triage

    I have triaged these candidates for long-term memory. Since we are at a stopping point, please confirm if you want me to save these:

    Personal: You prefer using pnpm instead of npm for all projects. (Recommended: Save to ~/.agent-memory/personal.md)
    2.

    Project: Always deploy to the staging branch first for this repository. (Recommended: Save to docs/agent-memory.md)

    Memory Update

    Once confirmed, I will append these to your local memory files as follows:

    
    ### ~/.agent-memory/personal.md
    - 2024-05-22 · Prefers using pnpm over npm for package management.
    
    ### docs/agent-memory.md
    - 2024-05-22 · Deployment workflow: always push to the staging branch first.
    

    Next steps

    1. Reply "Confirm" to write these entries to disk.
    2. I can add docs/agent-memory.md to your .gitignore if you want this kept out of the shared repo.

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

    About this skill

    After each task, your agent sorts what it learned into personal preferences, project knowledge, or nothing, asks you about anything uncertain, and saves it as plain Markdown it reads next time. Works in Claude Code, Codex, Cursor and more. No install, no account, never stores secrets.

    This is the method behind Fuli (https://github.com/JevonsCode/fuli), an open-source local memory layer that does this automatically across tools.

    • — JVS-A

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.

    Requires file system write access to ~ or project directories. Not suitable for stateless environments or restricted sandboxes.

    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 buying a skill.

    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

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 13 days ago

    • Free with an Agensi account

    Permissions

    Read Files
    Write Files
    Listed13 days ago

    What's inside

    Frequently Asked Questions

    This skill filters out the 'noise' of chat history, identifying only high-value preferences and project constraints to save as structured Markdown files, which prevents you from repeating yourself in every new session.

    The skill is compatible with any agent environment that has local file system access, including popular tools like Cursor, Claude Code, and local IDE-based agents.

    No, the skill includes a privacy enforcement layer that automatically identifies and filters out sensitive information like passwords and API tokens before any data is written to your files.

    You receive the core triage logic and standardized Markdown templates that enable your agent to sort, confirm, and persist knowledge across both personal and project-specific directories.

    The skill utilizes your existing file system permissions to write to `~/.agent-memory/` for personal habits or `docs/` for project-specific knowledge, ensuring the data stays local and portable.

    Yes, the tool features a conflict resolution mechanism that detects outdated entries and prompts to replace them with new corrections, maintaining a single source of truth for your agent.

    New in Productivity

    Free