Can I run long-running predictions with this skill?

Yes, while the 'wait_for' parameter supports up to 60 seconds of synchronous waiting, you can handle longer tasks asynchronously using webhooks or by polling the prediction list.

Do I need a Replicate API key to use this skill?

Yes, you must connect your Replicate account using an API token through the Composio MCP setup to authorize predictions.

Does this skill support uploading local files for model input?

Yes, the REPLICATE_CREATE_FILE tool allows you to upload base64-encoded file content, making it possible to provide images or documents as inputs for model predictions.

How do I determine the correct inputs for a specific Replicate model?

Use the REPLICATE_MODELS_GET tool to retrieve the 'openapi_schema', which provides a detailed breakdown of the required input keys and data types for that specific model.

What is Replicate Automation for Claude?

It is a specialized skill that allows Claude to interact with Replicate's API to run and manage machine learning models for images, text, audio, and video.

Replicate Automation

Name: Replicate Automation
Author: ComposioHQ

byComposioHQ

•

47,561

•

Data Science & ML

Automates the execution and management of Replicate AI models for image generation, LLM inference, and multimedia processing.

Replicate Automation provides a comprehensive set of tools for interacting with the Replicate API directly from Claude. It allows users to browse public models, inspect complex input/output schemas, upload necessary assets like images or audio, and trigger predictions for everything from Flux image generation to Llama-3 text completion. By handling the complexities of model versions, synchronous waiting, and file management, this skill enables developers to integrate state-of-the-art AI models into automated workflows and applications without leaving the Claude environment.

Key Features

0147,561 GitHub stars

02Upload and manage binary files for models requiring image or document inputs.

03Inspect model schemas and READMEs to ensure precise input parameter mapping.

04Track prediction history and list available model versions for stable deployments.

05Support for synchronous execution (up to 60s) and asynchronous webhook-based predictions.

06Execute predictions on public models including LLMs, image generators, and audio/video tools.

Use Cases

01Automating media transformation tasks like audio transcription, video upscaling, or background removal.

02Integrating specialized open-source LLMs and multi-modal models into agentic workflows.

03Building automated image generation pipelines using models like Flux.1 or SDXL.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add composiohq/awesome-claude-skills replicate-automation

For use in Claude.ai and ChatGPT

Download Skill