Stash, developed by Collective Returns, Inc., is an MCP server tool for the mcp platform. It is built with Python and distributed as a Python package from its GitHub repository. Stash provides powerful context to enhance the AI coding experience. It connects agentic apps to its Issue Resolution Assistance, operating as a set of Model Context Protocol tools. This server is designed to offer advanced dataset queries and automated insights for various applications.

What does Stash connect to?

The Stash MCP Server connects to a "Stash instance" to link agentic apps with Stash's Issue Resolution Assistance. It functions as an MCP server, providing Model Context Protocol tools to enrich the AI coding experience. This connection supplies context from Stash, enabling agentic applications to access and utilize crucial information for their operations.

Core features and tools

The primary callable tool is get_all_performers, supporting advanced filtering by criteria like favorites_only, country, ethnicity, eye_color, hair_color, measurements, piercings, tattoos, height_cm, and weight. Modifiers for comparison types such as EQUALS, GREATER_THAN, and BETWEEN are available. Resource URIs include stash://performer/all, stash://performer/{name}, stash://performer/country/{country}, stash://performer/ethnicity/{ethnicity}, and stash://performer/stats. A scene listing capability with filters is also present.

Who is it for?

Stash targets AI coding agents such as Cursor, Claude, and GitHub Copilot, alongside developers. It is also for users seeking programmatic access to advanced dataset queries or automated insights for research, library curation, or personal discovery. This includes applications like performer/scene analysis and generating personalized recommendations. Skip this if you do not require programmatic access to a Stash instance.

How to get started

The server authenticates using an API key passed as an environment variable. The setup requires configuring a mandatory STASH_API_KEY environment variable.

Use cases

• Enhanced Issue Resolution for AI Agents: Connects AI agents to Stash's Issue Resolution Assistance, providing context like documentation, past issues, and code locations for accurate task completion.
• Programmatic Querying and Analysis of Stash Instances: Allows users to query and analyze a Stash instance with high-precision filters for performer/scene analysis and personalized recommendations.
• Automated Context Provisioning for AI Development: Supplies AI agents with critical information, aiding understanding of legacy codebases and building new components.

Stash is an MCP server tool that provides valuable context and query capabilities for AI agents and developers interacting with Stash instances. It supports advanced filtering for data analysis and assists AI agents in complex development tasks. The required API key ensures secure access to its features. Try using the get_all_performers tool with advanced filters for dataset analysis.