ModelRunner Agent Skills MCP server
Commit Nov 12, 2026Enables AI assistants to run multimodal models and generate App Store screenshots or iOS demo assets using the ModelRunner inference API.
Also known as modelrunner/agent-skills
- Stars
- 0
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 12, 2026
- License
- MIT
What is ModelRunner Agent Skills MCP?
ModelRunner Agent Skills enables AI assistants to run multimodal models and generate App Store screenshots or iOS demo assets using the ModelRunner inference API.
Summary generated from the project's README
About ModelRunner Agent Skills MCP server
Generating marketing visuals and running multimodal models from within your coding environment is what agent-skills enables. Developers using Claude Code, Cursor, OpenAI Codex, or GitHub Copilot can use these portable skills to execute complex tasks involving image, video, audio, 3D, and text through the ModelRunner inference API. The server connects to the remote ModelRunner MCP at mcp.modelrunner.run and requires a ModelRunner account with an active balance and OAuth authorization.
The toolkit includes a general modelrunner skill for finding and running specific models based on live schemas and pricing. It also features specialized tools for mobile developers, such as app-store-screenshots for creating cohesive marketing sets with device frames and ASO headlines, and ios-demo-assets for generating bulk images that load directly into Xcode Assets.xcassets and the iOS Simulator via xcrun simctl addmedia.
What you can do with it
- 1Create App Store marketing setsTransform raw application captures into a consistent set of branded screenshots with device mockups and headlines.
- 2Generate iOS simulator assetsProduce bulk demo images and automatically load them into the Xcode assets folder and the iOS Simulator library.
- 3Execute multimodal model tasksIdentify the correct model for a specific image, video, or audio task and run it while tracking costs.
Written from the project's README. Every use case is checked against a quote from it.
README
View on GitHubModelRunner Agent Skills
Portable AI-assistant skills for running any AI model and generating real product visuals —
App Store screenshots and iOS demo assets — powered by ModelRunner, the unified AI inference
API. Authored once as SKILL.md and built for Claude Code, OpenAI
Codex, Cursor, and GitHub Copilot.
Each skill calls ModelRunner's public MCP server (
mcp.modelrunner.run) — one API for image, video, audio, 3D, and text that you can use for anything else at modelrunner.ai »
Status: v1.2.1 — published and live on the Claude Code plugin marketplace as
modelrunner/agent-skills (add it with the command below). Three skills across four
tool targets: a general modelrunner skill plus two built from the
ModelRunner case studies.
Sample outputs
Real, unretouched results from the two case studies these skills are built from — every image below was generated through the ModelRunner MCP.
app-store-screenshots — raw app captures → a cohesive, on-brand store set (device frames,
generated backgrounds, ASO headlines), kept consistent via an anchor→style-reference pass:
ios-demo-assets — bulk, on-spec demo images loaded straight into Xcode's Assets.xcassets
and the Simulator's Photos library:
What's inside
| Skill | What it does |
|---|---|
modelrunner |
The general one: find the right model for any image, video, audio, speech or 3D task, read its live schema and price, run it, and return the outputs. Covers uploads, batches, run tagging and cost. |
app-store-screenshots |
Turns raw app captures into a cohesive set of App Store / Google Play marketing screenshots — device mockups, on-brand backgrounds, and ASO-aware headlines. Uses the modelrunner/app-store-screenshot-composer wrapper and the anchor→style-reference pattern for a consistent set. Case study » |
ios-demo-assets |
Bulk-generates on-spec demo images and loads them into Xcode's Assets.xcassets and the iOS Simulator's Photos library (xcrun simctl addmedia). Case study » |
Prerequisites
Every skill calls the ModelRunner MCP. Create a ModelRunner account with a small balance at
modelrunner.ai, then connect the server at https://mcp.modelrunner.run/mcp
and authorize with your account when prompted (OAuth).
In Claude Code, installing a plugin below wires the MCP for you; for other tools the MCP config
ships in dist/<tool>/.
Maintaining an open-source project? ModelRunner gives active open-source projects free monthly API credits — apply at modelrunner.ai/oss-program.
Install
Claude Code
/plugin marketplace add modelrunner/agent-skills
/plugin install modelrunner@modelrunner
/plugin install app-store-screenshots@modelrunner
/plugin install ios-demo-assets@modelrunner
Any Agent Skills tool (npx skills)
Installs the SKILL.md files into Claude Code, Codex, Cursor and other agents that read the
Agent Skills format. It does not configure the MCP server, so connect
https://mcp.modelrunner.run/mcp in your tool as well:
npx skills add modelrunner/agent-skills
OpenAI Codex CLI
SKILL.md is read natively. Copy the skills and add the MCP server:
cp -R dist/codex/skills/* ~/.codex/skills/
cat dist/codex/config.toml.snippet >> ~/.codex/config.toml
Cursor
Cursor reads the same Agent-Skills SKILL.md. Copy the generated .cursor/ into your project:
cp -R dist/cursor/.cursor /path/to/your-project/
GitHub Copilot (VS Code)
Copilot has no skill format, so each skill ships as a /-invokable prompt file:
cp -R dist/copilot/.github dist/copilot/.vscode /path/to/your-project/
Then run /modelrunner, /app-store-screenshots or /ios-demo-assets in Copilot Chat.
How it's built
SKILL.md is the single source of truth (read natively by Claude Code, Codex, and Cursor — the
"Agent Skills" open standard). A dependency-free build script emits each
tool's packaging into dist/:
npm run build # node scripts/build.mjs
dist/claude/— plugin marketplace (.claude-plugin/marketplace.jsonis at the repo root; plugin bodies +.mcp.jsonhere)dist/codex/—SKILL.mdskills + aconfig.tomlMCP snippetdist/cursor/—.cursor/skills/+.cursor/mcp.jsondist/copilot/—.github/prompts/*.prompt.md+.vscode/mcp.json
The name/URLs baked into the manifests come from package.json + mcp/modelrunner.json. Prefer a
turnkey multi-tool sync instead of the built-in script? rulesync
and ruler can emit the same targets from a
canonical source.
License
MIT · Built by ModelRunner — the unified AI inference API.
ModelRunner Agent Skills MCP questions
What can the modelrunner skill do?
It finds the appropriate model for image, video, audio, speech, or 3D tasks, reads its schema and price, and executes the task.
How do I generate App Store screenshots?
Use the app-store-screenshots skill to turn raw app captures into a cohesive set of marketing screenshots with device mockups and backgrounds.
Can these skills be used for iOS development?
Yes, the ios-demo-assets skill generates demo images and can load them directly into Xcode's Assets.xcassets and the iOS Simulator's Photos library.
Does ModelRunner offer support for open-source projects?
Yes, active open-source projects can apply for free monthly API credits.