# klavis-ai/klavis **Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/klavis-ai-klavis).** 5,640 stars · 531 forks · Python · apache-2.0 ## Links - GitHub: https://github.com/Klavis-AI/klavis - Homepage: https://www.klavis.ai/ - awesome-repositories: https://awesome-repositories.com/repository/klavis-ai-klavis.md ## Topics `agents` `ai` `ai-agents` `api` `developer-tools` `discord` `function-calling` `integration` `llm` `mcp` `mcp-client` `mcp-server` `oauth2` `open-source` ## Description Klavis is a platform for managing Model Context Protocol (MCP) servers and providing sandboxed environments where AI agents can safely interact with external tools and services. It functions as an integration framework that orchestrates MCP server instances, exposes tools and resources for AI agents, and isolates agent interactions from production data through horizontally scalable sandbox environments. The platform distinguishes itself through its ability to generate long-horizon agentic tasks that simulate realistic tool-use workflows with live SaaS applications and production MCP servers. It includes an agentic task generator that creates multi-step coding and tool-use challenges for training and evaluating AI agents, with deterministic world state seeding and programmatic outcome verification using tests, rubrics, and LLM-based judges. Klavis also optimizes agent context windows by structuring step-by-step interactions into efficient execution paths, and supports intent-driven action discovery that maps natural language queries to server capabilities without semantic search. The platform provides capabilities for deploying over 100 prebuilt MCP server integrations with OAuth support via Docker or CLI commands, managing server instances through REST APIs, and authenticating external services through OAuth flows and API key management. It supports creating, resetting, and seeding isolated sandbox environments with predefined datasets, and allows agents to use MCP tools in sandboxed environments as if they were live services. ## Tags ### Part of an Awesome List - [AI Agent Frameworks](https://awesome-repositories.com/f/awesome-lists/ai/ai-agent-frameworks.md) — Manages MCP server instances, execution trajectories, and sandboxed environments for AI agent workflows. ### Artificial Intelligence & ML - [MCP Server Management](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol/mcp-server-management.md) — Manages MCP server instances through a REST API for creation, configuration, and lifecycle control. ([source](https://api.klavis.ai/v1/mcp-server/instance)) - [Client Registrations](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol/mcp-server-management/mcp-server-connections/mcp-client-disconnections/client-registrations.md) — Provides the mechanism to register MCP server endpoints with AI assistant clients like Cursor and Claude Code. ([source](https://www.klavis.ai/docs/quickstart)) - [Sandbox Environments](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol/mcp-server-management/sandbox-environments.md) — Provides isolated sandbox environments where AI agents can safely interact with MCP tools and external services. - [AI Agent Servers](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-agent-servers.md) — Serves as an AI agent server that exposes integration schemas and tools via the Model Context Protocol. ([source](https://api.klavis.ai/v1/mcp-server/strata)) - [Intent-Based Discoveries](https://awesome-repositories.com/f/artificial-intelligence-ml/artificial-intelligence-tooling/agent-and-tool-integrations/mcp-server-integrations/action-function-discoveries/intent-based-discoveries.md) — Maps natural language queries to server categories and actions without using semantic search. - [Context Window Optimizations](https://awesome-repositories.com/f/artificial-intelligence-ml/context-window-optimizations.md) — Optimizes context windows by structuring agent-environment interactions into efficient execution paths. ([source](https://cdn.jsdelivr.net/gh/klavis-ai/klavis@main/README.md)) - [Agent Output Verifiers](https://awesome-repositories.com/f/artificial-intelligence-ml/custom-evaluation-judges/deterministic-evaluators/agent-output-verifiers.md) — Evaluates agent outputs using deterministic tests, rubrics, and LLM-based judges for granular reward signals. ([source](https://www.klavis.ai/)) - [Programmatic Configurations](https://awesome-repositories.com/f/artificial-intelligence-ml/language-model-orchestration/mcp-server-configurations/programmatic-configurations.md) — Ships SDKs and REST APIs for programmatically creating and configuring MCP server instances. ([source](https://cdn.jsdelivr.net/gh/klavis-ai/klavis@main/README.md)) - [Data Exposures](https://awesome-repositories.com/f/artificial-intelligence-ml/mcp-servers/data-exposures.md) — Creates MCP servers that expose connected accounts and data to clients through unique URLs and access tokens. ([source](https://www.klavis.ai/docs/quickstart)) - [Model Context Protocol Servers](https://awesome-repositories.com/f/artificial-intelligence-ml/model-context-protocol-servers.md) — Exposes tools and resources for AI agents to interact with external systems and services via MCP. - [Workflow Simulation Generators](https://awesome-repositories.com/f/artificial-intelligence-ml/tool-using-agents/workflow-simulation-generators.md) — Generates long-horizon agentic tasks that simulate realistic tool-use workflows with live SaaS applications. ([source](https://www.klavis.ai/)) - [Trajectory-Based Agent Optimization](https://awesome-repositories.com/f/artificial-intelligence-ml/trajectory-based-agent-optimization.md) — Captures and structures step-by-step agent interactions to reduce token consumption and optimize context windows. - [MCP Server Connections](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol/mcp-server-management/mcp-server-connections.md) — Connects to remote MCP servers for CRM and project management to optimize AI agent context windows. ([source](https://www.klavis.ai/docs/mcp-server/overview)) - [OAuth-Enabled Deployments](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol/mcp-server-management/oauth-enabled-deployments.md) — Deploys and configures prebuilt MCP servers with OAuth support for third-party service integration. - [Long-Horizon Coding Task Generators](https://awesome-repositories.com/f/artificial-intelligence-ml/generative-code-models/long-horizon-coding-task-generators.md) — Creates multi-step coding challenges for training reinforcement learning models. ([source](https://www.klavis.ai/)) - [Execution Trajectory Managers](https://awesome-repositories.com/f/artificial-intelligence-ml/step-based-schedulers/step-execution-engines/execution-step-controllers/execution-trajectory-managers.md) — Captures and structures step-by-step agent interactions to optimize context window usage. ([source](https://www.klavis.ai/docs)) ### Development Tools & Productivity - [Agent Sandboxes](https://awesome-repositories.com/f/development-tools-productivity/agent-sandboxes.md) — Provides isolated, horizontally scalable environments for AI agents to interact with external services safely. - [MCP Server Integrations](https://awesome-repositories.com/f/development-tools-productivity/ai-assistant-integrations/mcp-server-integrations.md) — Exposes tools and resources through MCP servers so AI agents can interact with external systems. - [Prebuilt MCP Deployments](https://awesome-repositories.com/f/development-tools-productivity/ai-assistant-integrations/mcp-server-integrations/prebuilt-mcp-deployments.md) — Deploys over 100 prebuilt MCP servers with OAuth support via Docker or CLI commands. ([source](https://cdn.jsdelivr.net/gh/klavis-ai/klavis@main/README.md)) - [Intent-Based Discoveries](https://awesome-repositories.com/f/development-tools-productivity/api-discovery-tools/server-capability-discovery/intent-based-discoveries.md) — Implements intent-driven discovery that maps natural language queries to server capabilities without semantic search. ([source](https://www.klavis.ai/docs/concepts/strata)) - [Isolated Evaluation Environments](https://awesome-repositories.com/f/development-tools-productivity/isolated-execution-environments/isolated-evaluation-environments.md) — Provisions dedicated, horizontally scalable sandbox environments that isolate agent interactions from production data. - [Agent Sandbox Provisioners](https://awesome-repositories.com/f/development-tools-productivity/isolated-execution-environments/isolated-evaluation-environments/agent-sandbox-provisioners.md) — Provisions dedicated, horizontally scalable sandbox environments for AI agent testing. ([source](https://www.klavis.ai/docs/concepts/sandbox)) - [MCP Connectivity](https://awesome-repositories.com/f/development-tools-productivity/sandboxed-execution-environments/sandbox-configuration/mcp-connectivity.md) — Lets agents use standard MCP tools against sandboxed environments as if they were live services. ([source](https://www.klavis.ai/docs/concepts/sandbox)) - [Agentic Task Generators](https://awesome-repositories.com/f/development-tools-productivity/workflow-task-generators/agentic-task-generators.md) — Generates long-horizon coding and tool-use tasks for training and evaluating AI agents. ### DevOps & Infrastructure - [Training Task Environments](https://awesome-repositories.com/f/devops-infrastructure/container-orchestration/container-runtimes/runtime-configuration-interfaces/docker-socket-orchestrators/docker-target-configurators/docker-container-deployments/docker-container-execution/training-task-environments.md) — Ships coding tasks inside Docker containers with deterministic tests for reproducible agent training runs. ([source](https://www.klavis.ai/)) - [Task Environment Packages](https://awesome-repositories.com/f/devops-infrastructure/container-orchestration/container-runtimes/runtime-configuration-interfaces/docker-socket-orchestrators/docker-target-configurators/docker-container-deployments/docker-container-execution/task-environment-packages.md) — Packages coding challenges and tests inside Docker containers for reproducible agent training runs. ### Security & Cryptography - [Third-Party Authentication Providers](https://awesome-repositories.com/f/security-cryptography/third-party-authentication-providers.md) — Completes OAuth flows and manages API keys to authorize integrations with third-party applications. ([source](https://www.klavis.ai/docs/quickstart)) - [OAuth Authentication APIs](https://awesome-repositories.com/f/security-cryptography/oauth-authentication-apis.md) — Completes OAuth flows and manages API keys to authorize integrations with third-party applications. - [OAuth Integration Managers](https://awesome-repositories.com/f/security-cryptography/oauth-integration-managers.md) — Handles OAuth authentication flows and API key management for third-party service connections. ### Testing & Quality Assurance - [World State Seedings](https://awesome-repositories.com/f/testing-quality-assurance/deterministic-state-testing/world-state-seedings.md) — Loads predefined datasets into sandbox environments to create reproducible test scenarios for agents. - [Long-Horizon Workflow Simulations](https://awesome-repositories.com/f/testing-quality-assurance/local-agent-simulation/agent-to-tool-simulations/long-horizon-workflow-simulations.md) — Generates long-horizon agentic tasks that simulate realistic tool-use workflows with live SaaS applications. - [Agent Output Verifications](https://awesome-repositories.com/f/testing-quality-assurance/validation-verification/input-validation/agent-input-and-output-validators/automated-assertion-validators/action-result-validators/outcome-assertions/agent-output-verifications.md) — Evaluates agent outputs using deterministic tests, rubrics, and LLM-based judges for granular reward signals. - [Deterministic World State Seeders](https://awesome-repositories.com/f/testing-quality-assurance/deterministic-state-testing/deterministic-world-state-seeders.md) — Loads predefined datasets into sandbox environments for reproducible agent testing. ([source](https://www.klavis.ai/docs/concepts/sandbox)) - [Sandbox State Verifiers](https://awesome-repositories.com/f/testing-quality-assurance/software-testing/testing-frameworks/test-frameworks/assertions-and-validation/test-case-definitions/parametrized-test-cases/output-comparison-judges/sandbox-state-verifiers.md) — Exposes sandbox state for comparison against ground-truth expectations to judge task completion. ([source](https://www.klavis.ai/docs/concepts/sandbox)) ### Software Engineering & Architecture - [On-Demand Action Executions](https://awesome-repositories.com/f/software-engineering-architecture/custom-action-handlers/on-demand-action-executions.md) — Provides on-demand execution of server actions with provided parameters through its MCP server platform. ([source](https://www.klavis.ai/docs/concepts/strata))