Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Selective Context Memory

    1

    Most AI coding agent setups dump every past decision, bug fix, and session note into context on every prompt.

    Free

    Security scanned

    About this skill

    Most AI coding agent setups dump every past decision, bug fix, and session note into context on every prompt. That wastes tokens and degrades output quality as your project grows.

    Selective Context Memory fixes this with a simple, file-based approach:

    • Topic-scoped memory files — decisions.md, api.md, debug.md — one per topic, not one per date
    • An index in your constitution file that tells the agent WHICH file to read for the current task
    • Automatic session logging via hooks, so the agent can self-orient on resume without you re-explaining context
    • Append-only entries: date, what happened, one-line why

    Works with Claude Code, Cursor, GitHub Copilot, Gemini CLI, and any agent that reads markdown instruction files.

    What's included

    • Complete SKILL.md with setup instructions and usage rules
    • Starter templates for memory/decisions.md, memory/api.md, memory/debug.md
    • Ready-to-use Claude Code hooks config (SessionStart, SessionEnd, PreCompact)
    • Copy-paste snippet for any agent's constitution/rules file
    • Scaling guide for solo projects through large codebases

    No dependencies. No database. No vector store. Just plain markdown files your agent reads and writes.

    Requires agent filesystem access. Initial manual setup of directory structure and index file mapping is necessary. Best suited for projects where agents can execute shell hooks.

    Open docs

    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

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 7 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed7 days ago

    What's inside

    Frequently Asked Questions

    This skill prevents 'context bloat' by forcing your agent to only load specific, task-relevant logs (like API docs or debug history) instead of reading the entire project history, which saves tokens and improves response accuracy.

    The skill is compatible with any AI agent that has local filesystem access, including Claude Code, Cursor, Copilot, and Gemini CLI, as it relies on a standardized Markdown indexing system.

    You will receive the core indexing logic, a suite of standardized memory templates (decisions, debug, API), and the configuration instructions to integrate the system into your agent's primary instruction file.

    Setup involves initializing a designated memory directory and mapping your task types to specific log files in your agent's configuration file, a process that typically takes less than five minutes.

    Yes, because it uses a standard Markdown-based architecture, you can easily customize the number of memory files or the specific topics to fit the unique needs of your project or workflow.

    While the agent automates the reading and appending of logs, it does require a one-time manual setup of the directory and the initial index file to establish the memory boundaries.

    New in Productivity

    Free