Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Optimization-Loop

    1

    Autonomous loop that iteratively modifies, evaluates, and selects the best version of any text resource — skills, prompts, or campaigns — using a modify-measure-keep/discard cycle.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    1 useSecurity scanned
    Optimization-Loop

    Optimization-Loop

    Example session with this skill installed

    Customer: John Smith
    Email: john@example.com
    API Key: sk-123456789

    • Read your context and instructions
    • Compiled the optimization-loop

    Customer: [PERSON_1]
    Email: [EMAIL_1]
    API Key: [API_KEY_1]

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

    About this skill

    Auto-Optimize is an autonomous optimization loop for text-based resources such as agent skills, prompts, and campaign copy. Inspired by Andrej Karpathy's autoresearch methodology, it applies the same principle that works for training models — modify, measure, keep or discard, repeat — to the domain of text optimization.

    The skill operates on files stored in a dedicated targets directory. Each optimization cycle makes exactly one targeted change, evaluates the result using either live metrics or simulated LLM-judge scoring, and logs the outcome to a results file. If the change improves the score, it's kept and becomes the new baseline. If it performs the same or worse, it's discarded via a hard git reset. This ensures the optimization path is always traceable, reversible, and never accumulates unverified changes.

    The evaluation method adapts to the target type: skills are scored on task success rate through agent testing, prompts through simulated conversation scoring across multiple criteria, and campaigns through real click and conversion data. When live metrics are unavailable, the skill uses delegate_task with an evaluator LLM to simulate and score conversations based on per-target evaluation criteria.

    Key constraints keep the process grounded: only one change per experiment, a simplicity criterion where removing code that performs equally is always kept, no bundling of multiple ideas, and changes must be large enough to matter but small enough to review. The loop runs continuously until interrupted by a human, and it never asks for permission to continue — it just optimizes.

    • Requires git for state management.
    • High token usage during long loops.
    • LLM-judge scoring can be subjective.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 5 months ago

    Permissions

    Terminal / Shell

    File Scopes

    .md
    Listed5 months ago

    What's inside

    Frequently Asked Questions

    This skill automates the 'trial and error' of prompt engineering and script refinement by applying a modify-measure-discard cycle, ensuring your agents or campaigns improve based on data rather than intuition.

    The skill supports two modes: it can integrate live performance data (like CTR) or use an LLM-judge to simulate interactions and score them against your specific success criteria.

    Optimization-Loop is designed to work with any text-based resource, including agent system prompts, marketing copy, and code-based skills, provided they are stored in the designated targets directory.

    It utilizes a git-based 'hard reset' mechanism, meaning if an experimental change fails to improve the score, the skill immediately reverts the file to its last known best state.

    The skill follows a strict 'one change per experiment' rule and records every iteration in a results log, allowing you to audit the entire evolutionary path of your resource.

    You receive the full Optimization-Loop skill package, documentation on setting up your targets directory, and example evaluation templates for both automated and human-in-the-loop scoring.

    New in Agents & Orchestration