Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    autocritic

    1

    Enforce a self-critical, collaborative workflow that stops agents from making unapproved changes or outputting low-quality code.

    Free

    1 useSecurity scanned
    autocritic

    autocritic

    Example session with this skill installed

    Add a basic authentication layer to my Express API. I have the files open.

    • Read your context and instructions
    • Compiled the autocritic

    I found that you have an existing Express setup in src/app.js but no middleware for handling authentication or user sessions.

    I plan to

    1. Install bcryptjs and jsonwebtoken for secure hashing and token management.
    2. Create a src/middleware/auth.js file to verify JWTs.
    3. Update the POST /login route to validate credentials and return a token.
    4. Add the auth middleware to the protected /api/v1/dashboard route.

    What do you think?

    [Refinement Note: Switched from jsonwebtoken.sign to a dedicated utility function to ensure consistent expiration and algorithm settings. Added a check for the JWT_SECRET environment variable.]

    Next steps

    • Approve the plan to begin file creation.
    • Provide the preferred token expiration duration.
    • Confirm if you have a .env file ready for the secret key.

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

    About this skill

    Agents often produce code too quickly, ignore architectural trade-offs, or clutter the terminal with verbose commentary. This skill enforces a partner-like collaboration model where the agent investigates first, checks in before any modification, and performs an invisible self-review to ensure high-quality output without the "theater" of fake tests.

    What it does

    • Ambiguity classification categorizes requests to decide whether to stop and ask for clarity or proceed with best-practice assumptions.
    • Pre-modification check-ins ensure the agent never edits files or creates scaffolding without first sharing a plan for user approval.
    • Invisible self-review runs an internal senior-engineer perspective shift to catch edge cases, security flaws, and logic gaps before delivery.
    • State persistence maintains a .project/ directory to track session goals, architectural decisions, and deferred uncertainties.
    • Refinement reporting provides a concise note only when substantive corrections were made during the internal review process.

    How it works

    1. Analyze the request for ambiguity and either ask up to three targeted questions or proceed to investigation.
    2. Investigate the codebase or environment freely without needing permission to read files or run diagnostics.
    3. Check in by presenting a concrete action plan and waiting for user confirmation before modifying any project state.
    4. Execute and review the task internally, applying domain-specific lenses (Code, Content, or Architecture) to prune errors.

    Frameworks & tools

    This is a model-agnostic behavioral layer designed for terminal-based AI environments like Claude Code, Gemini CLI, Aider, and Cursor. It works across any programming language or stack by governing the agent's interaction logic rather than specific syntax.

    Why this beats prompting it yourself

    Prompting an agent to "be careful" or "check in" is often ignored during long sessions. This skill provides a rigid execution framework with specific ambiguity routing and a mandated internal checklist that forces a perspective shift, preventing the agent from agreeing with its own initial mistakes.

    Use cases

    • Preventing agents from refactoring the wrong files in large, complex repositories.
    • Ensuring architectural decisions are approved before a multi-file migration begins.
    • Stripping out "AI fluff" and fake test blocks from CLI-based coding sessions.
    • Maintaining a durable log of technical debt and deferred questions in the .project/ folder.

    Known limitations

    The skill requires the agent to have file-system access to maintain project state. Performance may degrade if the model lacks an extended reasoning trace for the internal review step.

    Requires file system access for .project/ state. Effectiveness relies on the model's ability to perform an internal perspective shift. Not intended for simple, one-off chat queries.

    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
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    It solves the problem of agents rushing into edits without permission, preventing unapproved refactors and low-quality code by mandating a "plan-first" approval step and an internal senior-engineer review.

    This is a model-agnostic behavioral layer that integrates with terminal-based environments like Claude Code, Cursor, Aider, and Gemini CLI, regardless of the underlying programming language.

    Autocritic requires an environment where the agent has file-system access, as it uses a dedicated `.project/` directory to track architectural decisions and session goals across restarts.

    Unlike standard prompts that agents often ignore over time, Autocritic provides a rigid execution framework with specific ambiguity routing that forces a perspective shift to catch its own logic gaps.

    You will receive the core behavioral logic files and a configuration guide for implementing the workflow in your preferred AI terminal environment.

    The skill includes a 'Refinement Reporting' feature that ensures you only see a concise summary when substantive corrections were made, keeping your terminal clean of unnecessary 'AI fluff.'

    New in Workflow Automation