{"categories":[{"label":"Application Frameworks","url":"https://skillfed.io/packages/category/software-development-libraries-application-frameworks/3"}],"enrichment":{"capability":"Python SDK for the Plato platform that manages sessions, environments, and agents for browser automation and task evaluation within Plato VMs.","skillfed_tags":["plato-platform","environment-provisioning","agent-framework"],"use_cases":["Create isolated test environments for evaluating agent behavior against predefined test cases with automatic mutation logging and evaluation scoring.","Spin up simulator instances (Gitea, Kanboard, etc.) for development or testing, then execute shell commands and retrieve state mutations programmatically.","Build custom agents by subclassing BaseAgent and publish them to Plato's package registry for use in platform VMs.","Run browser automation workflows by combining Plato environments with the computer-use agent for interactive task completion.","Monitor environment state and reset individual simulators during test runs without recreating the entire session."],"what_it_does":"Plato SDK v2 is a Python client library for the Plato platform, a service that provisions and manages isolated virtual environments (simulators) for testing, evaluation, and automation tasks. It provides both async and sync interfaces to create sessions, spin up environments like Gitea or Kanboard, execute commands within them, and evaluate task completion against predefined test cases. The SDK handles session lifecycle (creation, reset, cleanup), exposes public URLs for browser access, and integrates with Plato's agent system for custom automation workflows.\n\nThe package is designed for developers building evaluation frameworks, automated testing pipelines, or custom agents that need to interact with ephemeral environments. It depends on a substantial stack including aiohttp for async HTTP, boto3 for cloud operations, litellm for LLM integration, pydantic for validation, and opentelemetry for observability. Configuration is environment-based (API key and base URL), and agents can be published to Plato's PyPI via CLI commands.","worth_installing":"Yes, if you are building on the Plato platform or need a managed environment provisioning and evaluation framework. The SDK is actively maintained (released 2 days ago), has low install friction, carries no known vulnerabilities, and is MIT-licensed. The 29 runtime dependencies are substantial but expected for a platform SDK. Install only if you have a Plato API key and are committed to the platform's architecture; it is not a general-purpose automation library."},"id":"plato-sdk-v2","links":{"html":"https://skillfed.io/packages/plato-sdk-v2","md":"https://skillfed.io/packages/plato-sdk-v2.md","pypi":"https://pypi.org/project/plato-sdk-v2/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-12","license_spdx":"MIT","license_treatment":"permissive","name":"plato-sdk-v2","python_support":"supports_current","summary":"Python SDK for the Plato API"},"popularity":{"monthly_downloads":395602,"position":6980,"tier":"top_15000"},"security":{"n_vulnerabilities":0},"version":"2.127.1"}