instagram-play

The Instagram Play skill empowers AI agents to operate the Instagram mobile application autonomously, providing a comprehensive solution for social media management on Android devices. This skill is designed for power users, social media managers, and individuals looking to offload repetitive engagement tasks to an intelligent agent. By leveraging the OpenPocket framework, the agent can interact with the app as a human would—navigating menus, scrolling through content, and triggering specific UI components without requiring constant human oversight.

The agent handles complex workflows such as creating Instagram Stories, managing direct messages, and engaging with Reels or feed posts. It is particularly effective for users who need to maintain an active presence on the platform while minimizing manual interaction time. The agent reads the device screen in real-time to make informed decisions, ensuring it correctly identifies UI elements even when the app updates its interface. It supports advanced features like batch action processing for typing and sending messages, as well as failure handling mechanisms to manage session expirations or upload retries.

• Automated story creation: Navigate to profile, access the camera/gallery, select media, apply overlays, and publish efficiently.
• Feed and content engagement: Scroll through the home feed, like posts, comment, and share content via DMs.
• Reels navigation: Interactive control for swiping through short-form video content and engaging with creators.
• Direct Messaging: Read, reply, and manage private conversations directly through the app interface.
• Intelligent navigation: Uses core UI identifiers (home, search, reels, profile) to move between application sections reliably.
• Robust error handling: Automatically handles retries for media uploads, manages popups, and requests human authentication via OAuth if the session expires.

Usage notes and practical constraints:

• Ensure the Instagram app is installed and logged in on the target Android device or emulator.
• The agent relies on visual cues from the screen; maintain a stable internet connection for smooth media loading.
• Always verify account permissions if managing multiple Instagram profiles.
• The skill includes a failure handling protocol: if media fails to upload, the agent will attempt a single retry before notifying the user.
• Use in conjunction with human-in-the-loop workflows for sensitive tasks like initial account linking via OAuth.
• Requires active accessibility or ADB-based input permissions on the host device to execute tap, swipe, and text input actions.