Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Gemini Image Gen

    1

    Automate Google Gemini image generation and full-sized downloads using verified browser-skill workflows.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Gemini Image Gen

    Gemini Image Gen

    Example session with this skill installed

    Use Gemini to generate a high-resolution 16:9 cinematic landscape image based on my prompt and save the full-sized file to my desktop.

    • Read your context and instructions
    • Compiled the gemini image gen
    • Generated the image

    I have successfully initiated the Gemini image generation process. The environment check confirmed the Gemini session, and the prompt was injected into the rich-text editor. I monitored the mat-progress-spinner to ensure the 20-second generation cycle completed before triggering the high-resolution download.

    Gemini Generation Result

    MetricValue
    StatusCompleted
    Original Resolution2816 × 1536 (16:9)
    File NameGemini_Generated_Image_20260920_1430.jpg
    Target PathC:/Users/shizh/Desktop/landscape_cinematic.jpg
    Generation Time42 seconds
    VerificationVerified (>100KB, Full-size confirmed)

    The image has been successfully moved from the Downloads directory to your specified desktop location. The session has been closed, and the browser tab has been returned.

    Next steps

    • Review the generated image at the desktop path to ensure it meets your aesthetic requirements.
    • If the model filtered the prompt, I can use soften_prompts.py to rewrite the description for a successful second attempt.
    • To generate a series of similar images, provide a folder of reference images to initiate the batch processing script.

    Gemini_Generated_Image_20260920_1430.jpg

    PNG · 1536×1024

    Generated

    Example file from a real run - the skill writes it into your workspace.

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

    About this skill

    Automating high-resolution image generation in Google Gemini often fails due to complex rich-text editors, file upload limitations, and misleading UI spinners. This skill provides a battle-tested automation workflow using browser-skill (bsk) to handle full-sized original image generation, reliable multi-image uploads, and robust download detection.

    What it does

    • Automated Image Prompting bypassing bsk fill issues using evaluate and insertText injections.
    • High-Resolution Retrieval ensuring full-sized original downloads (up to 2816x1536) instead of low-res thumbnails.
    • Reference Image Management implementing a clipboard-based upload strategy to handle up to 7 reference images per task.
    • Precise Completion Tracking using specific DOM polling that avoids the common "loading spinner" false-completion pitfall.
    • Batch Processing including scripts for task orchestration, dependency tracking, and breakpoint resumption.

    How it works

    1. Session Setup initializes bsk status and borrows the Gemini tab via session start.
    2. Text Injection focuses the rich-textarea and executes insertText to avoid editor re-renders.
    3. Reference Upload utilizes Windows clipboard CF_HDROP and Ctrl+V commands for reliable image attachment.
    4. Polling & Verification monitors mat-progress-spinner and lastImgs counts to confirm the model response is complete.
    5. Download Capture records a directory baseline and polls for new files to circumvent extension attribution timeouts.

    Frameworks & tools

    Built for Python 3.13 and browser-skill (bsk). It uses ctypes for clipboard manipulation and raw struct parsing for image dimension validation without external imaging libraries.

    Why this beats prompting it yourself

    Manual Gemini automation via standard browser skills frequently times out or grabs low-quality blob URLs. This skill provides verified selectors and timing logic that handle 10-minute generation cycles and security filter bypasses that standard agents miss.

    Use cases

    • Generating high-resolution marketing assets with specific aspect ratios (9:16, 4:3, 16:9).
    • Batch producing images based on a set of consistent reference character photos.
    • Automating the creation of large image datasets for downstream design pipelines.
    • Rescuing stuck generation tasks using pre-built "harvest" and "soften" scripts.

    Known limitations

    Requires an active, logged-in Gemini session in a browser reachable by bsk. Native Bash commands like sleep and ls are unavailable, requiring Python-based directory polling.

    Requires logged-in Google Gemini session. Limited to Windows for clipboard-based uploads. Does not support PIL for image verification; uses internal struct parsing for dimensions.

    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

    Money-back guarantee

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

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files
    Environment Variables
    Listed19 days ago

    What's inside

    Frequently Asked Questions

    Standard automation often pulls low-resolution thumbnails; this skill uses a specialized polling and capture method to ensure you retrieve the full-sized original files (up to 2816x1536) directly from the Gemini interface.

    The skill is specifically designed for Python 3.13 environments utilizing the `browser-skill` (bsk) framework and requires an active, logged-in Google Gemini session in your browser.

    Yes, the skill includes a clipboard-based injection strategy that allows you to reliably upload and manage up to 7 reference images per generation task without triggering UI errors.

    Standard 'loading' detection often fails in Gemini's complex editor; this skill uses specific DOM polling and directory baselining to confirm the generation is truly finished before attempting a download.

    You receive the core automation workflows, specialized scripts for handling text injection and multi-image uploads, and batch processing utilities for managing high-volume tasks.

    While it automates the input and retrieval process, the user must provide the logged-in browser session, as the skill does not bypass Google's account authentication or security protocols.

    New in Agents & Orchestration