For Claude Code · Last updated August 28, 2026
How to add image and video tools to Claude Code
Claude Code writes code and runs your terminal, but it has no image, video, or audio model of its own. Install the AnyCap skill and CLI once and it can generate and inspect media in the same session, then save the results into your project.
Paste this into Claude Code
Claude Code follows the install guide, adds the skill and CLI, and asks you to log in once. The manual setup is further down.
Set up AnyCap using anycap.ai/install.txt, then show me what you can do with images and video.What does AnyCap add to Claude Code?
Claude Code keeps the code and terminal work. Each row below is one CLI command away once AnyCap is installed.
| Capability | Claude Code alone | With AnyCap | |
|---|---|---|---|
| Image generation | No image output from terminal | Generate visuals and mockups via anycap image generate | Image Generation page |
| Video generation | No video tooling in CLI workflow | Create walkthroughs and clips via anycap video generate | Video Generation page |
| Music generation | No music generation in CLI workflow | Create background tracks through the AnyCap music runtime | Music Generation page |
| Image understanding | No unified vision runtime | Read screenshots, diagrams, and visual references | Image Understanding page |
| Video analysis | Requires separate provider setup | Inspect recordings from the same CLI | Video Analysis page |
| Audio understanding | No unified audio analysis runtime | Transcribe and analyze audio through one runtime | Audio Understanding page |
| Web search | Built-in WebSearch depends on provider and permissions | Search the web from the same capability layer | Web Search page |
| Grounded web search | Source retrieval depends on WebSearch and WebFetch access | Run grounded search when the answer needs citations | Grounded Web Search page |
| Web crawl | No reusable crawl runtime | Crawl pages and extract content from one CLI | Web Crawl page |
| Social profile lookup | No structured social profile data | Read Instagram and X profiles with follower and post counts | Social lookup page |
| Social post lookup | No structured post data from a social URL | Fetch an Instagram post or reel, or an X post, by URL | Social lookup page |
| Social account feeds | No account feed retrieval | List recent Instagram posts and reels or X posts from an account | Social lookup page |
| X post search | No structured X post search | Search X posts by keyword or search expression | Social lookup page |
| Drive storage | No shared asset storage layer | Store outputs with public URLs in AnyCap Drive | Drive page |
| Drive share links | No managed share links for outputs | Share Drive files with optional password and expiry | Drive page |
| Project snapshots | No portable project handoff | Package a project into one password-protected snapshot link | Drive page |
| Page hosting | No built-in page publishing surface | Publish simple pages through AnyCap Page | Pricing page |
| Page versions and rollback | No versioned publishing history | Keep every Page deploy as a version and roll back | Pricing page |
| Visual annotation review | Text-only human feedback | Collect marked-up feedback on images, URLs, video, and audio | Human-in-the-loop guide |
| Collaborative whiteboard | No shared diagram surface | Open an Excalidraw board from Mermaid and update it live | Human Interaction skill |
| One auth flow | Per-provider credential management | One login across the capability stack | Get Started page |
Which task are you adding?
Each guide covers one task: the install path, which model to pick, and the exact command Claude Code runs.
Image generation →
Mockups, hero images, and icons, saved straight into the repo.
anycap image generate --model seedream-5 --prompt "a minimal SaaS dashboard on a light background" -o dashboard-hero.pngVideo generation →
Product clips and walkthrough b-roll from a text prompt or a still image.
anycap video generate --model veo-3.1 --prompt "a developer typing in a dark terminal, slow push-in"Video analysis →
Ask a question about a screen recording and get notes Claude Code can act on.
anycap actions video-read --file ./recording.mp4 --instruction "List every error message shown on screen"Screenshot review →
Point it at a screenshot or diagram and ask what is wrong.
anycap actions image-read --file ./bug-screenshot.png --instruction "What UI issue do you see?"Claude Opus 5.5 →
Opus 5.5 writes text only. Add image, video, and audio output.
How do you set it up by hand?
Three commands, run once in the terminal where Claude Code runs.
1. Install the skill
npx -y skills add anycap-ai/anycap -a claude-code -yThis teaches Claude Code how to discover and invoke the AnyCap capability runtime.
2. Install the CLI
curl -fsSL https://anycap.ai/install.sh | shThe CLI is a single binary with no runtime dependencies — it runs directly in Claude Code's terminal session.
3. Log in and verify
anycap login && anycap statusAfter authentication, Claude Code can move across image, video, and vision capabilities without new credentials or dashboard detours.
For a full walkthrough, see the install guide.
Why does it fit Claude Code?
It runs where Claude Code runs
The CLI is one binary with no runtime dependencies. Claude Code installs it, logs in, and calls it in the same terminal session.
Results land in your project
Commands return a file path or a URL. Claude Code can save the file into the repo, embed it in Markdown, or pass it to the next step.
One login instead of three
Image, video, and vision providers each want their own key and SDK. AnyCap puts them behind one login.
FAQ
Can Claude Code analyze videos?
Claude Code does not ship with a built-in video-analysis runtime. With AnyCap installed, it can call the video-read action on a remote video or uploaded local recording, ask a focused question, and use the returned findings in the same terminal workflow. Use the Video Analysis capability page for supported inputs, CLI examples, representative output, and limitations.
Can Claude Code generate images on its own?
No. Claude Code focuses on code reasoning, planning, and terminal execution. It has no built-in image generation runtime. AnyCap adds that capability through one skill install and one CLI, so Claude Code can produce visuals without leaving your terminal session or wiring a separate image provider.
Why use AnyCap instead of wiring providers directly?
Wiring a separate image API, a video API, and a vision API means managing three sets of credentials, SDKs, and authentication flows. AnyCap consolidates those into one CLI and one login that Claude Code can invoke directly from your terminal. One skill install replaces multiple provider integrations.
Does AnyCap replace Claude Code?
No. AnyCap is not an agent. It is a capability runtime that runs alongside Claude Code. You keep Claude Code for code, planning, and terminal execution, and add the image, video, and vision tools it does not ship with. AnyCap extends Claude Code — it never competes with it.
What is the fastest path to add tools to Claude Code?
Just tell Claude Code 'help me install anycap.ai' in natural language. It discovers the skill, installs the CLI, authenticates, and is ready to generate images or video immediately. No commands to memorize. If you prefer manual control, you can also install the skill and CLI yourself in three steps.
Does AnyCap work inside the Claude Code terminal?
Yes. The AnyCap CLI is a single binary with no external dependencies. It runs directly in Claude Code's terminal session, sends API requests to the AnyCap server, and returns file paths or URLs that Claude Code can use in subsequent steps, embed in code, or save to your project directory.
Last updated August 28, 2026