Support

PromptLens support

Find installation steps, provider setup notes, permission explanations, screenshot selection shortcuts, and common error guidance.

Installation

Download from GitHub Releases

Use the latest release package from GitHub Releases. Chrome Web Store listing is planned.

Load from source

  1. Download or clone the repository.
  2. Open Chrome and go to chrome://extensions.
  3. Enable Developer mode.
  4. Click Load unpacked.
  5. Select the extension folder.

Configure your AI provider

PromptLens does not provide an AI model or API key. Open the options page and configure:

  • AI Base URL: an HTTPS OpenAI-compatible endpoint, for example https://api.openai.com/v1.
  • API Key: your provider key.
  • Model: a model that supports vision input.
  • Template: the output structure you want PromptLens to request.

Local development endpoints may use http://localhost or http://127.0.0.1.

Image read permission

Remote image analysis may require optional site access so PromptLens can read image URLs from websites or access the API origin you configure. This is an optional host permission, not a required install-time site permission.

You can refuse remote image read permission and use screenshot selection instead.

Screenshot selection

Use Alt+Shift+S on Windows/Linux or Option+Shift+S on macOS to start screenshot selection. You can also use the context menu action.

Screenshot selection is useful for blob: images, hotlink-protected images, embedded canvases, or pages where remote image reads fail. Press Esc or click Cancel to exit selection mode.

Common errors

Missing API configuration

Open the options page and save your Base URL, API key, and vision-capable model name.

API authentication failed

Check whether the API key is valid for the provider and model you configured.

Model does not support vision input

Choose a model that accepts image input through an OpenAI-compatible vision request format.

Remote image cannot be read

Grant optional image read permission for the site, or use screenshot selection for the visible region.

Image is too large

Adjust image processing preferences in options, or use screenshot selection to capture a smaller region.

SVG is not supported

Use screenshot selection to capture the rendered SVG area as a raster image.

Blob images require screenshot selection

blob: URLs are page-local and may not be readable from the extension context. Select the visible page region instead.

Contact

The default support channel is GitHub Issues. Please include your browser version, PromptLens version, provider type, model name, and the exact error message when reporting a problem.

Do not post API keys, private image data, or sensitive prompts in public issues.