Record for AI agents

Record for agent a specialized recording mode in Loom that turns your selected screen recordings into high-fidelity instructions for AI agents. By combining your voice instructions with on-screen interactions, visited links, and technical metadata, Loom creates a structured “agent prompt” that agents like Rovo or Claude can use to execute tasks.

For example, you can record a video explaining a UI change while navigating between a Figma design and a live website. Loom captures the visual context and your verbal intent, then generates a prompt for you to review and share with an agent or automatically create work items in Jira for agents to act on. This eliminates the need to manually write long prompts or copy-paste multiple screenshots to give an agent context.

Create an agent prompt

You can use Record for agent mode to capture complex instructions and automatically generate a prompt for AI agents to follow.

Option 1: Record using the Loom Chrome extension

1. Open the Loom Chrome extension.

2. Select the Generate tab from the recorder menu.

3. Select Record for agent from the list of enhanced recording modes.

4. Select Start recording.

5. Verbally describe the actions you want the agent to take while navigating to the relevant screens, buttons, or links.

6. Select the Stop button when you finish your instructions.

7. View the generated Action Plan in the Generate tab on the Loom share page.

8. If you have an Atlassian-managed account, select Create in Jira to automatically turn specific actions into work items for agents to act on.

   • Alternatively, copy the prompt to paste it directly into a third-party agent such as Claude.

Option 2: Record using the Loom Desktop App

1. Open the Loom desktop app (Version 2.0 or later)

2. Select the Record for agent option in the recorder menu.

3. Click the Record for agent button to start recording.

4. As you navigate and desribe the actions you want an agent to take, use the on-screen selector tools to guide the agent:

   • Selection tool: Click on specific UI elements or containers on your screen to pin precise visual context for the agent.

   • Magic pen tool: Draw temporary freehand annotations directly on your screen to emphasize visual shapes or areas to the agent.

5. Click Stop when you finish your recording.

Share the agent prompt

Once you stop recording, you will be automatically taken to the Generate tab of the Loom share page of your video. On the share page, you can:

1. View the generated agent prompt.

2. Click Send to agent at the top right of the agent prompt panel to hand off your instructions. From the dropdown, you can:

   • Select your preferred agent: Loom will automatically launch the tool with your prompt pre-filled (or copy the prompt to your clipboard for terminal/CLI tools).

   • Open in Jira: If you have an Atlassian-managed account, automatically turn the prompt into a work item for agents to act on.

   • Copy as prompt: Copy the raw markdown prompt to your clipboard to use it anywhere else, such as terminal tools or custom workflows.

Record for agent settings and capabilities

This section describes the specific data captured and the limitations of the Record for agent mode during its open beta phase.

Data capture

When you record Record for agent mode, Loom captures enhanced metadata to ensure AI agents have full context. This includes:

• Transcript: A full text record of your spoken instructions.

• Keyframes: Visual snapshots of the screens you shared, including the elements selected with the Selection tool and freeform drawings made with the Magic pen tool.

• Interaction events: Data on where you clicked during the recording.

• Visited links: A log of the URLs you navigated to while recording.

• Console and network logs (Chrome extension): console.log, warn, error, navigation messages, and network request/response metadata.

Recording limits

Record for agent videos have specific constraints to optimize them for AI processing:

• Time limit: Recordings are limited to a maximum of five minutes. A countdown timer appears during the recording to help you track your time.

• Camera: The camera is automatically turned off in Record for agent mode to focus the agent's attention on the screen content.

Agent prompt

The agent prompt summarizes your video into discrete, executable tasks optimized for AI agents. Each task in the Agent prompt typically includes:

• A concise title and bulleted instructions.

• A transcript quote for verbal context.

• Visual references or keyframes from the recording.

• Timestamps indicating when the action was discussed.

This feature should be used in accordance with Atlassian’s Acceptable Use Policy.