AnyCap for Devin Desktop — Add Image, Video & Media to Your Agent IDE

Add image generation, video production, and media publishing to Devin Desktop agents with AnyCap CLI. Zero custom code. Works with any Devin Desktop agent in minutes.

by AnyCap

Devin Desktop Kanban board dashboard showing AI agent task cards in Queued, In Progress, and Done columns — clean corporate blue and white UI

Devin Desktop gives you the Agent Command Center, Kanban-managed agents, and a full IDE in one place. What it doesn't include is any way for your agents to create images, generate video, or publish media. If your workflow ever needs a hero image, a product demo clip, or a batch of social graphics, your agents have to stop and hand off to a human.

AnyCap's CLI changes that. Install it once, and every agent running inside Devin Desktop — Devin, your custom ACP agents, or any other agent in the workspace — can generate images, produce video, synthesize audio, search the web, and publish files to a CDN. No additional configuration per agent.


What AnyCap Adds to Devin Desktop

Image Generation

Generate product mockups, hero images, UI assets, social graphics, and illustrations from any agent session. Access 15+ leading image models — including FLUX Pro, Stable Diffusion XL, Ideogram, and Recraft — from a single tool call, with no per-model API setup required.

Video Production

Generate demo videos, walkthroughs, social clips, and explainer content using Veo 3, Kling, Runway, Wan, and more — without leaving Devin Desktop. Your agents can produce video as naturally as they produce code.

Audio and Music

Generate background music, sound effects, voiceovers, and audio assets for your projects. AnyCap supports speech synthesis, music generation, and audio processing tools — all callable from agent sessions.

Web Search and Crawl

Give your agents real-time web access. AnyCap's search and crawl tools let Devin Desktop agents fetch current information, read competitor pages, analyze live documentation, and act on fresh data without manual intervention.

File Publishing and Sharing

Upload generated assets to cloud storage and receive shareable CDN links instantly. AnyCap handles file persistence so generated media is immediately usable in code, documentation, pull request descriptions, or external systems.


How to Set Up AnyCap for Devin Desktop

Step 1: Install the AnyCap CLI

npm install -g anycap
anycap login

That's the full setup. AnyCap is a CLI tool. Devin Desktop agents run shell commands as part of their work — once AnyCap is installed and authenticated on the system, any agent in your workspace can invoke it directly.

Step 2: Verify it works

Open any agent session in Devin Desktop and type:

"Use AnyCap to generate a test image — a simple gradient square."

The agent runs anycap image generate via shell, gets back a file path and CDN link, and returns it in the session output.


Why AnyCap Works Well Here

Devin Desktop agents — whether Devin, a custom ACP agent, or any other agent running in the workspace — can execute shell commands as part of their tasks. AnyCap is a CLI tool, so it fits naturally into that model: install it once on the system, and agents can call anycap commands whenever a task requires image generation, video production, audio synthesis, web search, or file publishing.

Agent tool calls return local file paths and CDN URLs, so generated media fits directly into chained workflows. An agent building a feature can generate a hero image, get back the CDN URL, and reference it in a PR description or pass it to another agent via Spaces without any human step in between.


What This Looks Like in Practice

A Devin agent building a new landing page writes the HTML, CSS, and JS, then generates the hero image and a short demo video in the same session, and opens a PR with all the assets included. That used to require handing the task back to a human after the code was done.

Marketing content works the same way: an agent reads your changelog, generates a LinkedIn carousel image set, writes the copy, and outputs everything as a publish-ready package. For QA documentation, an agent can capture UI states, generate annotated screenshots of key flows, write the docs, and link everything in your wiki — no manual screenshot step.

For client presentations, agents can generate polished mockups and a brief demo video from a project brief, which gives you something concrete to send before the feature is built.


What Is Devin Desktop?

Devin Desktop is Cognition's next-generation IDE, built on the Windsurf foundation and redesigned around agent management at scale. It evolved from the premise that serious engineering teams are no longer pair-programming with one agent at a time — they are delegating tasks to multiple agents, reviewing their progress, and deciding what ships.

Key capabilities Devin Desktop brings:

  • Agent Command Center — A Kanban board showing all running local and cloud agents, their status, output, and blockers in real time
  • Spaces — Shared context layers that let multiple agents working on related tasks stay synchronized
  • ACP protocol — Any Agent Client Protocol-compatible agent can run inside Devin Desktop, not just Devin
  • Full VS Code compatibility — Extensions, keybindings, LSPs, and terminals all work exactly as before

AnyCap is a CLI tool that works wherever agents can run shell commands. Install it once, and every agent in your Devin Desktop workspace can generate and publish media as naturally as it writes code.


Get Started

Install AnyCap and connect to Devin Desktop

Read the integration docs