Shimmering Obsidian includes two distinct screenshot hotkeys: one that captures a screen region and extracts its text via OCR, appending the result to a Markdown note, and one that saves the raw image as a PNG directly into your vault and embeds it in a runningDocumentation Index
Fetch the complete documentation index at: https://mintlify.com/chrisgrieser/shimmering-obsidian/llms.txt
Use this file to discover all available pages before exploring further.
Images.md log. Both are triggered without leaving your current application.
OCR Screenshots
The OCR screenshot feature lets you select any region of your screen and automatically convert whatever text appears there into Markdown — useful for capturing text from PDFs, slide decks, terminal windows, or any other source where copy-paste is unavailable. Trigger: invoke the hotkey assigned to the OCR screenshot action in Alfred Preferences. This works just like macOS’s built-in⌘ ⇧ 4 shortcut — a crosshair cursor appears and you drag to select the region you want.
After the selection is confirmed, Tesseract processes the image and the extracted text is appended to your OCR screenshot file (by default OCR-Screenshots.md in your vault root), then opened in Obsidian.
Requirement: Install Tesseract
The OCR feature depends on Tesseract, an open-source OCR engine. If you use Homebrew, install it with:tesseract-lang package includes language data files for dozens of languages beyond English.
Language Configuration
For best OCR accuracy, tell Tesseract which languages to expect. Set theocr_languages workflow variable to the appropriate language code or codes. Use + to combine multiple languages — for example, eng+deu for English and German.
You can look up the correct code for your language in the Tesseract language data file list.
Output File
The workflow appends OCR’d text to a single file. By default this isOCR-Screenshots.md in your vault root. Configure ocr_screenshot_file in the workflow variables to redirect output to a different path (absolute or ~-prefixed paths are supported).
Configuration Reference
| Variable | Description |
|---|---|
ocr_languages | Tesseract language code(s) to use, e.g. eng or eng+deu. |
ocr_screenshot_file | Absolute or ~-relative path to the file where OCR text is appended. Defaults to {vault}/OCR-Screenshots.md. |
Image Screenshots
The image screenshot feature captures a screen region as a PNG and saves it directly into your vault, then embeds the image in a dedicatedImages.md note at your vault root.
Trigger: invoke the hotkey assigned to the image screenshot action in Alfred Preferences (separate from the OCR hotkey). Select the region you want to capture the same way you would with ⌘ ⇧ 4.
Filename and Save Location
Every captured screenshot is named automatically using the date and time of capture:screenshot_path workflow variable to the desired path (absolute or ~-prefixed). The folder is created automatically if it does not exist.
Embedding in Images.md
After saving the PNG, the workflow appends an Obsidian image embed to Images.md in your vault root:
Images.md does not yet exist, it is created automatically. If it already exists, the embed is appended to the end — so every screenshot you take accumulates in one place in chronological order.
Configuration Reference
| Variable | Description |
|---|---|
screenshot_path | Folder where PNG files are saved. Defaults to {vault-path}/screenshots/. Accepts absolute or ~-prefixed paths. |
Granting Screen Recording Permission
Both screenshot features require Alfred to have permission to record the screen. macOS will prompt you the first time automatically, but if you missed the prompt or denied it, you can grant permission manually:Enable Alfred
Find Alfred in the list and toggle it on. If Alfred is not listed, click the + button and add it from your Applications folder.