The Action Library

Displays the current output in the runner.

Prompts the user for text, a number, or a URL.

Displays a modal message that needs acknowledgement.

Adds a note that does not run.

Stores a value under a reusable name — leave the parameter empty to capture the previous output. Reference it later as {name}.

Reads a value saved earlier in the shortcut.

Runs a branch only when a condition matches. Needs nested steps — not executable yet.

Repeats the nested actions a fixed number of times. Needs nested steps — not executable yet.

Runs the nested actions once for each item in a list. Needs nested steps — not executable yet.

Presents named options and continues from the selected option.

Lets the user select one or more items from a list.

Gets the first, last, random, or indexed item from the previous output.

Counts items, characters, words, or lines.

Pauses execution for a number of seconds.

Provides plain text to the next action.

Extracts readable text from the current input.

Joins a list of text values with a separator.

Finds and replaces text in the previous output.

Divides text into a list by lines, spaces, commas, or custom text.

Converts text to uppercase, lowercase, or title case.

Removes extra spaces and blank lines from text.

Reads text aloud using the system speech engine when available.

Translates text through the configured translation provider.

Calls the OpenAI Responses endpoint with a prompt and optional previous output.

Analyzes an image URL or file reference through a vision-capable OpenAI model.

Generates speech audio from text through the OpenAI audio speech endpoint.

Transcribes an audio file through the OpenAI audio transcription endpoint.

Translates an audio file into English through the OpenAI audio translations endpoint.

Creates an image from a prompt through the OpenAI image generation endpoint.

Creates vector embeddings from text through the OpenAI embeddings endpoint.

Prepares a custom OpenAI API call with an endpoint path and JSON body.

Calls a local Zallama (OpenAI-compatible) server with a prompt and optional previous output.

Sends a system prompt and user message to a local Zallama (OpenAI-compatible) chat endpoint.

Creates vector embeddings from text through a local Zallama (OpenAI-compatible) embeddings endpoint.

Searches images, videos, or audio from a selected folder or library.

Gets the most recent images, videos, or audio items.

Saves generated or selected media to a folder.

Resizes an image to a target width, height, or percentage.

Crops an image to a rectangle or aspect ratio.

Plays an audio file or stream.

Turns speech in an audio file into text.

Opens the default mail client with recipient, subject, and body prefilled.

Creates a calendar event through a configured calendar provider.

Creates a task in the configured reminders provider.

Launches an installed app by name or path.

Creates a note in the configured notes provider.

Runs a GET or POST request and passes along the response.

Opens a URL in the default browser.

Returns platform, architecture, and app environment details.

Shows a local in-app notification.

Copies text (or the previous output) to the system clipboard.

Saves text output to a file (relative paths land in the vault).

Lists files and folders from a directory (relative paths resolve in the vault).

Creates a folder (relative paths land in the vault).

Converts supported text, HTML, or images into a PDF.

Extracts searchable text from a PDF.

Calls another saved Zeshort shortcut.

Runs a shell command (cmd on Windows, sh elsewhere) and passes along its output. Use {variable} to insert saved values.

Evaluates a math expression — use {variable} to insert saved values.

Stops execution immediately.

Stops execution and returns the current value.

Summarizes text using the configured AI provider.