Works with the AI tools you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseOpenClawOpenClaw+21 more

    CodeGuardian

    1

    Enforce surgical atomic diffs and prevent accidental code deletion during agent-led refactoring.

    $19

    Secure checkout via Stripe

    30-day refund guarantee

    Converts to your local currency at checkout

    Security scanned
    CodeGuardian

    CodeGuardian

    Example session with this skill installed

    Add a try-except block to the process_payment method in payment_service.py to catch and log Stripe exceptions.

    • Read your context and instructions
    • Compiled the codeguardian

    🔍 CODE INTEGRITY VERIFIED. ENFORCING DIFF OUTPUT ONLY.

    Code Modification

    @@ -42,5 +42,9 @@ class PaymentService:
     def process_payment(self, amount, currency):
-        response = stripe.Charge.create(amount=amount, currency=currency)
-        return response
+        try:
+            response = stripe.Charge.create(amount=amount, currency=currency)
+            return response
+        except stripe.error.StripeError as e:
+            logger.error(f"Stripe payment failed: {str(e)}")
+            raise PaymentError("Provider communication failure")

    Next steps

    1. Run the local test suite to verify the PaymentError is caught by the controller.
    2. Check the application logs to ensure the error string is formatted correctly.

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

    What you get

    Prevent agents from deleting functions during large file refactors.Stop models from using lazy placeholders like 'rest of code here'.Generate immediately executable unified diffs for IDE integration.Enforce immediate code output by bypassing verbose planning steps.

    About this skill

    Coding agents often destroy working code by hallucinating full-file rewrites or omitting "unchanged" blocks. This skill forces surgical precision, ensuring every modification is an atomic diff that preserves existing functions and imports.

    What it does

    • Enforces atomic diffs by restricting output to unified diff blocks with hunk headers instead of full-file rewrites.
    • Prevents code deletion by flagging and rebuilding patches if existing functions or imports disappear without instruction.
    • Blocks planning demagogy by skipping multi-step explanations and confirmation trees in favor of immediate execution.
    • Eliminates placeholders by treating any # ... rest unchanged ... or ellipsis as a hard failure.
    • Structural validation runs a pre-output check for bracket balance, indentation layers, and dependency mapping.

    How it works

    1. Analyze boundaries to identify the exact line ranges of the target block and map surrounding dependencies.
    2. Verify integrity by checking the proposed change against the current file state to detect accidental regressions.
    3. Execute immediately by bypassing planning steps or "would you like me to" prompts.
    4. Emit unified diff using standard @@ headers, - removals, and + additions in a single code block.

    Frameworks & tools

    This skill is language-agnostic and works with any file-based codebase. It is compatible with IDE-based agents including Claude Code, Cursor, Codex CLI, Gemini CLI, and GitHub Copilot.

    Why this beats prompting it yourself

    Standard prompts often fail to stop agents from truncating large files when they get tired or run out of context window. This skill implements a strict failure-and-rebuild loop that treats code omission as a regression, ensuring you never lose a helper function to a lazy rewrite.

    Use cases

    • Injecting logging or telemetry into large, established source files without touching adjacent logic.
    • Performing localized bug fixes in modules where full-file regeneration is risky or slow.
    • Applying hotfixes in IDE environments where the model defaults to placeholders.
    • Adding single methods to classes without re-serializing the entire class structure.

    Known limitations

    This skill is not for initializing new repositories or generating boilerplate. It requires an existing file structure to generate diffs against.

    How to install

    Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.

    ~30 seconds
    1. 1

      Download the ZIP

      Free skills download straight away. Paid skills unlock right after purchase.

    2. 2

      Unzip into your skills folder

      Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.

    3. 3

      Ask your agent to use it

      Restart the agent if it was already running. It picks the skill up automatically - no config needed.

    Skills folder by agent

    Click the path to copy it. Create the folder if it does not exist yet.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean 2 days ago

    • Passed all security checks, Safe to install

    Listed2 days ago

    What's inside

    Frequently Asked Questions