Agent skill · software engineering · composiohq
Replicate Automation
Automate Replicate AI model operations -- run predictions, upload files, inspect model schemas, list versions, and manage prediction history via the Composio MCP integration.
Why this skill is useful
Adds executable scripts for managing Replicate AI model operations that aren't covered in public documentation.
What it needs
Requires composio installed locally. Requires replicate account access. About 3k tokens when loaded. Last updated 2026-07-24. 71,973 stars on the source repository.
What this skill does
Replicate Automation Automate your Replicate AI model workflows -- run predictions on any public model (image generation, LLMs, audio, video), upload input files, inspect model schemas and documentation, list model versions, and track prediction history. Toolkit docs: composio.dev/toolkits/replicate --- Setup 1. Add the Composio MCP server to your client: https://rube.app/mcp 2. Connect your Replicate account when prompted (API token authentication) 3. Start using the workflows below --- Core Workflows 1. Get Model Details and Schema Use REPLICATEMODELSGET to inspect a model's input/output schema before running predictions. Important: Each model has unique input keys and types. Always check the openapischema from this response before constructing prediction inputs. 2. Run a Prediction Use REPLICATEMODELSPREDICTIONSCREATE to run inference on any model with optional synchronous waiting and webhooks. Sync vs Async: Use waitfor (1-60s) for fast models. For long-running jobs, omit it and use webhooks or poll via REPLICATEPREDICTIONSLIST. 3. Upload Files for Model Input Use REPLICATECREATEFILE to upload images, documents, or other binary inputs that models need. 4. Read Model Documentation Use REPLICATEMODELSREADMEGET to access a model's README in Markdown format for detailed usage instructions. 5. List Model Versions Use REPLICATEMODELSVERSIONSLIST to see all available versions of a model, sorted newest first. 6. Track Prediction History and Files Use REPLICATEPREDICTIONSLIST to retrieve prediction history, and REPLICATEFILESGET/REPLICATEFILESLIST to manage uploaded files. --- Known Pitfalls Pitfall Detail --------- -------- Model-specific input keys Each model has unique input keys and types. Using the wrong key causes validation errors. Always call REPLICATEMODELSGET first to check the openapischema. File upload encoding REPLICATECREATEFILE requires base64-encoded content. Binary files treated as text (UTF-8) will fail with decode errors. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills composiohq/replicate-automation