{"categories":[{"label":"Python Modules","url":"https://skillfed.io/packages/category/software-development-libraries-python-modules/15"},{"label":"Artificial Intelligence","url":"https://skillfed.io/packages/category/scientific-engineering-artificial-intelligence/7"}],"enrichment":{"capability":"Stirrup is a lightweight Python framework for building LLM-powered agents that can execute code, search the web, process files, and call external tools with minimal scaffolding.","skillfed_tags":["agent-framework","llm-tools","async-first"],"use_cases":["Build an autonomous research agent that searches the web, fetches articles, and synthesizes findings into a report","Create a code-generation agent that writes, tests, and refines scripts based on user requirements","Develop a data analysis agent that processes files, executes analysis code, and generates visualizations","Set up a multi-tool agent that integrates with external services via MCP to automate complex workflows","Prototype a specialized agent by cloning the repository and customizing the core loop and tool set"],"what_it_does":"Stirrup is a lightweight agent framework designed to work with language models rather than impose rigid workflows on them. It provides a foundation for building autonomous agents that can execute code (locally, in Docker, or E2B sandboxes), search and fetch web content, call external tools via MCP servers, process files, and incorporate human feedback. The framework includes built-in context management to handle long conversations, multimodal support for images and video, and a flexible tool system based on Pydantic schemas.\n\nThe package ships with sensible defaults\u2014local code execution and web search tools\u2014but is fully customizable. You can use it as an installed package or fork it as a template for building specialized agents. It supports multiple LLM providers through OpenAI-compatible APIs, LiteLLM, or custom clients, and includes a skills system for modular, domain-specific instruction packages. The 11 runtime dependencies cover HTTP requests, image and video processing, schema validation, and retry logic.","worth_installing":"Yes. Stirrup is actively maintained, has low install friction, carries a permissive MIT License, and addresses a real need for a lightweight, unopinionated agent framework. Its 548 GitHub stars and recent activity suggest community traction. No known security vulnerabilities. Best suited for developers building agents who want flexibility over rigid frameworks."},"id":"stirrup","links":{"html":"https://skillfed.io/packages/stirrup","md":"https://skillfed.io/packages/stirrup.md","pypi":"https://pypi.org/project/stirrup/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-04","license_spdx":null,"license_treatment":"permissive","name":"stirrup","python_support":"supports_current","summary":"The lightweight foundation for building agents"},"popularity":{"monthly_downloads":241087,"position":8888,"tier":"top_15000"},"security":{"n_vulnerabilities":0},"version":"0.2.0"}