# langchain-ai/langchain-mcp-adapters

**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/langchain-ai-langchain-mcp-adapters).**

3,366 stars · 367 forks · Python · mit

## Links

- GitHub: https://github.com/langchain-ai/langchain-mcp-adapters
- Homepage: https://docs.langchain.com/oss/python/langchain/mcp
- awesome-repositories: https://awesome-repositories.com/repository/langchain-ai-langchain-mcp-adapters.md

## Topics

`langchain` `langgraph` `mcp` `python` `tools`

## Description

This project provides a translation layer and set of adapters designed to bridge AI agents with the Model Context Protocol. It functions as an integration layer that allows agents to operate as protocol-compliant servers and enables the conversion of protocol-based tools into formats compatible with agent frameworks and logic graphs.

The adapters facilitate tool interoperability by wrapping external protocol tools for use within agent workflows and exposing internal agent capabilities to any client implementing the Model Context Protocol. This creates a communication bridge that supports inter-agent discovery and coordination through a standardized protocol.

Beyond tool translation, the project covers the orchestration of agentic workflows, including stateful thread management, asynchronous execution, and real-time event streaming. It also incorporates support for OAuth 2.0 authentication, durable execution checkpointing, and the deployment of agents via Docker containers.

## Tags

### Artificial Intelligence & ML

- [Agentic Workflow Orchestration](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-architectures/orchestration-engines/ai-agent/agentic-workflow-orchestration.md) — Implements a system for managing the execution and state persistence of AI agents across synchronous and asynchronous runs.
- [Agent Communication Protocols](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-communication-protocols.md) — Implements the Model Context Protocol to expose assistants as compliant agents for seamless inter-agent interaction. ([source](https://docs.langchain.com/langsmith/agent-server-api/system/api-documentation.md))
- [Long-term Memory Stores](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-architectures/memory-management-systems/long-term-memory-stores.md) — Implements persistent storage mechanisms to retain context across multiple user sessions and threads. ([source](https://docs.langchain.com/langsmith/agent-server-api/system/api-documentation.md))
- [Recurring Agent Scheduling](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-architectures/orchestration-engines/ai-agent/runtime-execution-control/recurring-agent-scheduling.md) — Enables periodic agent invocations on a fixed timetable, supporting both isolated and state-sharing runs. ([source](https://docs.langchain.com/langsmith/agent-server-api/system/api-documentation.md))
- [Agent-to-Agent Communication](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-communication-protocols/agent-to-agent-communication.md) — Implements standardized interfaces and protocols to facilitate direct communication and coordination between different AI agents. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Agent Context Management](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-context-management.md) — Versions and organizes bundles of instructions and tools to ensure consistent agent behavior. ([source](https://docs.langchain.com/langsmith/context-hub.md))
- [Agent Deployment Management](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-deployment-management.md) — Provides comprehensive management of agent server deployments, including creation, updates, and status monitoring. ([source](https://docs.langchain.com/api-reference/deployments-v2/get-deployment.md))
- [Agent Deployment Runtimes](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-deployment-runtimes.md) — Provides a runtime environment that manages execution, state, and persistence for deployed agent graphs. ([source](https://docs.langchain.com/langsmith/components.md))
- [Agent State Persistence](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-state-persistence.md) — Executes assistants on specific threads to update and persist conversational state based on accumulated outputs. ([source](https://docs.langchain.com/langsmith/agent-server-api/system/api-documentation.md))
- [Assistant Metadata](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-capabilities-skills-tooling/agent-management/assistant-metadata.md) — Provides operations for retrieving and inspecting the configuration and metadata of specific assistant instances. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/get-assistant.md))
- [Agent Tooling Protocols](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-capabilities-skills-tooling/agent-tooling-protocols.md) — Implements standardized communication specifications to transform Model Context Protocol tools for use in agent frameworks. ([source](https://cdn.jsdelivr.net/gh/langchain-ai/langchain-mcp-adapters@main/README.md))
- [Agent Instantiation](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-orchestration-multi-agent/autonomous-agents/agent-orchestration-frameworks/agent-instantiation.md) — Implements the process of creating active agent instances from specific configuration settings and graphs. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/create-assistant.md))
- [Durable Execution Runtimes](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/agent-runtimes/agent-execution-runtimes/durable-execution-runtimes.md) — Automatically saves runtime state via checkpointing to enable recovery of agent processes from the exact point of failure.
- [Model Context Protocol](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-integration-interfaces/model-context-protocol.md) — Implements the Model Context Protocol to connect AI models to local data sources and external tools.
- [Multi-Protocol Tool Exposures](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-agent-integrations/ai-agent-tool-integrations/multi-protocol-tool-exposures.md) — Makes internal agent capabilities accessible as compliant tools for any client implementing the Model Context Protocol. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Assistant Configuration Updates](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-assistant-configurations/assistant-configuration-updates.md) — Enables modification of an assistant's name, description, and graph association via patch requests. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/patch-assistant.md))
- [External Tool Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/external-tool-integrations.md) — Provides connectors for bridging external tool schemas and communication protocols into agent logic graphs.
- [MCP Protocol Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/external-tool-integrations/mcp-protocol-integrations.md) — Wraps Model Context Protocol tools into compatible formats specifically for LangChain agent workflows and graphs. ([source](https://docs.langchain.com/langsmith/chat-prompt-engineering.md))
- [Protocol Tool Converters](https://awesome-repositories.com/f/artificial-intelligence-ml/function-to-tool-converters/protocol-tool-converters.md) — Wraps Model Context Protocol tools to make them accessible to external agent workflows and graphs.
- [Conversation Session Initializers](https://awesome-repositories.com/f/artificial-intelligence-ml/language-model-orchestration/conversation-management/conversation-session-initializers.md) — Implements mechanisms for bootstrapping conversation contexts and initializing stateful sessions. ([source](https://docs.langchain.com/langsmith/agent-server-api/threads/create-thread.md))
- [MCP Servers](https://awesome-repositories.com/f/artificial-intelligence-ml/mcp-servers.md) — Exposes AI agents as Model Context Protocol (MCP) servers so external clients can discover and use their capabilities. ([source](https://docs.langchain.com/langsmith/agent-server-api/mcp/mcp-get.md))
- [Model Context Protocol Implementations](https://awesome-repositories.com/f/artificial-intelligence-ml/model-context-protocol-implementations.md) — Implements a bridge that converts Model Context Protocol tools into compatible formats for agent frameworks.
- [Agent Execution Streaming](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-execution-streaming.md) — Provides real-time streaming of agent execution results from conversation threads to clients via open connections. ([source](https://docs.langchain.com/langsmith/agent-server-api/threads/join-thread-stream.md))
- [Agent Thread Command Interfaces](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-thread-command-interfaces.md) — Sends scoped protocol commands to agent threads via HTTP or WebSockets to trigger actions like starting runs. ([source](https://docs.langchain.com/langsmith/agent-server-api/streaming/protocol-v2-command.md))
- [Assistant Graph Retrieval](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-assistants/assistant-graph-retrieval.md) — Allows fetching the graph representation of a configured assistant, including depth-limited views. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/get-assistant-graph.md))
- [Repository Management](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-context-management/repository-management.md) — Manages the creation, retrieval, and deletion of agent skill and instruction repositories. ([source](https://docs.langchain.com/langsmith/context-hub.md))
- [Human-in-the-loop Workflows](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-orchestration-multi-agent/control-flow-and-workflows/human-in-the-loop-workflows.md) — Provides mechanisms for pausing agent execution to allow human review or approval of tool calls. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Agent Authentication](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-orchestration-multi-agent/security-and-auth/authentication-strategies/cli-agent/agent-authentication.md) — Handles OAuth 2.0 flows for agents, including pausing execution for user consent and token refreshing. ([source](https://docs.langchain.com/langsmith/agent-auth.md))
- [Agent Streaming Interfaces](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/agent-runtimes/agent-streaming-interfaces.md) — Streams agent execution events, including tool calls and assistant deltas, for a specific thread. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/create-run-stream-output.md))
- [Concurrency Managers](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/agent-runtimes/concurrency-managers.md) — Manages concurrent agent runs by controlling how new messages are queued or canceled while a run is in progress. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Event-Driven Model Streaming](https://awesome-repositories.com/f/artificial-intelligence-ml/event-driven-model-streaming.md) — Delivers real-time agent responses and execution updates using an event-driven streaming architecture.
- [Environment-Based Promotion](https://awesome-repositories.com/f/artificial-intelligence-ml/llm-application-orchestrators/environment-based-promotion.md) — Promotes versioned agent configurations and instructions through development, staging, and production environments. ([source](https://docs.langchain.com/langsmith/context-hub.md))
- [Agent Response Streams](https://awesome-repositories.com/f/artificial-intelligence-ml/multimodal-agent-capabilities/real-time-streaming/agent-response-streams.md) — Streams agent responses and tool results incrementally to clients in real time. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Remote Agent Management](https://awesome-repositories.com/f/artificial-intelligence-ml/remote-agent-management.md) — Supplies programmatic interfaces and remote connection tools to manage AI graphs as if they were local. ([source](https://docs.langchain.com/langsmith/components.md))
- [Remote Tool Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/remote-tool-integrations.md) — Unifies available capabilities from multiple remote tool servers into a single client interface. ([source](https://cdn.jsdelivr.net/gh/langchain-ai/langchain-mcp-adapters@main/README.md))
- [Stateless Run Executions](https://awesome-repositories.com/f/artificial-intelligence-ml/stateful-agent-orchestration/global-run-local-state/stateful-run-executions/stateless-run-executions.md) — Implements stateless graph invocations that return a run identifier immediately for background processing. ([source](https://docs.langchain.com/langsmith/agent-server-api/stateless-runs/create-background-run.md))
- [Tool Error Handling](https://awesome-repositories.com/f/artificial-intelligence-ml/tool-error-handling.md) — Provides logic for reporting external tool execution failures back to the model to allow for self-correction. ([source](https://cdn.jsdelivr.net/gh/langchain-ai/langchain-mcp-adapters@main/README.md))

### Development Tools & Productivity

- [Agent Graph Packaging](https://awesome-repositories.com/f/development-tools-productivity/agent-graph-packaging.md) — Provides a CLI to develop, test, and bundle agent computational graphs into deployable packages. ([source](https://docs.langchain.com/langsmith/components.md))
- [OAuth Provider Registrations](https://awesome-repositories.com/f/development-tools-productivity/dynamic-configuration-providers/dynamic-provider-registration/oauth-provider-registrations.md) — Implements automated registration of OAuth providers using auto-discovery to facilitate secure tool authentication. ([source](https://docs.langchain.com/api-reference/auth-service-v2/create-mcp-oauth-provider.md))
- [Local Development Servers](https://awesome-repositories.com/f/development-tools-productivity/local-development-servers.md) — Ships a lightweight local server with hot reloading and debugging for rapid agent testing. ([source](https://docs.langchain.com/langsmith/cli.md))
- [Deployment Revisions](https://awesome-repositories.com/f/development-tools-productivity/revision-histories/deployment-revisions.md) — Retrieves the history of deployment versions to track changes and manage release cycles. ([source](https://docs.langchain.com/api-reference/deployments-v2/list-revisions.md))
- [Scheduled Task Querying](https://awesome-repositories.com/f/development-tools-productivity/task-scheduling/scheduled-task-querying.md) — Filters and retrieves active periodic task configurations based on assistant IDs, thread IDs, or metadata. ([source](https://docs.langchain.com/langsmith/agent-server-api/crons/search-crons.md))
- [Thread Pruning](https://awesome-repositories.com/f/development-tools-productivity/thread-managers/thread-pruning.md) — Supports removing specific conversation threads and their accumulated outputs to optimize storage. ([source](https://docs.langchain.com/langsmith/agent-server-api/threads/delete-thread.md))
- [Agent Versioning](https://awesome-repositories.com/f/development-tools-productivity/version-management/agent-versioning.md) — Provides systems for tracking and selecting specific agent deployment versions at runtime. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/set-latest-assistant-version.md))

### DevOps & Infrastructure

- [Background Task Runners](https://awesome-repositories.com/f/devops-infrastructure/automation-orchestration/task-execution-frameworks/task-job-management/background-task-runners.md) — Runs agents asynchronously in the background without blocking the caller while maintaining state and lifecycle. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Deployment Management](https://awesome-repositories.com/f/devops-infrastructure/deployment-management.md) — Provides tools to list, delete, and monitor runtime and build logs for managed server deployments. ([source](https://docs.langchain.com/langsmith/cli.md))
- [Agent Run Scheduling](https://awesome-repositories.com/f/devops-infrastructure/recurring-job-scheduling/agent-run-scheduling.md) — Triggers periodic execution of agent graphs or assistants on new threads using cron schedules. ([source](https://docs.langchain.com/langsmith/agent-server-api/crons/create-cron.md))
- [Docker Image Building](https://awesome-repositories.com/f/devops-infrastructure/container-orchestration/container-runtimes/runtime-configuration-interfaces/docker-socket-orchestrators/docker-target-configurators/docker-container-deployments/docker-image-building.md) — Generates Docker images or standalone Dockerfiles based on project configuration for agent deployment. ([source](https://docs.langchain.com/langsmith/cli.md))
- [Custom Container Images](https://awesome-repositories.com/f/devops-infrastructure/container-orchestration/image-management-tools/custom-container-images.md) — Allows injecting custom shell commands into the container build process to install system dependencies. ([source](https://docs.langchain.com/langsmith/custom-docker.md))
- [Redeployments](https://awesome-repositories.com/f/devops-infrastructure/deployment-management/deployment-configuration/redeployments.md) — Triggers the redeployment of specific previous revisions of an agent server to production via API. ([source](https://docs.langchain.com/api-reference/deployments-v2/redeploy-revision.md))
- [Job Scheduling & Configuration](https://awesome-repositories.com/f/devops-infrastructure/job-scheduling-configuration.md) — Allows updating the execution schedule, timezone, and runtime configuration of recurring background tasks. ([source](https://docs.langchain.com/langsmith/agent-server-api/crons/update-cron.md))
- [Managed Agent Hosting](https://awesome-repositories.com/f/devops-infrastructure/managed-agent-hosting.md) — Builds and pushes server images to managed registries to create or update remote deployments. ([source](https://docs.langchain.com/langsmith/cli.md))
- [Run Cancellations](https://awesome-repositories.com/f/devops-infrastructure/workflow-run-management/run-cancellations.md) — Allows stopping active agent executions by targeting specific thread and run identifiers. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/cancel-runs.md))
- [Synchronous Run Executions](https://awesome-repositories.com/f/devops-infrastructure/workflow-run-management/synchronous-run-executions.md) — Supports triggering tasks within a thread while holding the connection open until the final output is produced. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/create-run-wait-for-output.md))

### Networking & Communication

- [AI Protocol Translation](https://awesome-repositories.com/f/networking-communication/cross-chain-messaging-protocols/protocol-translators/messaging-protocol-translation/ai-protocol-translation.md) — Transforms Model Context Protocol tool definitions into formats compatible with agent frameworks and logic graphs.
- [JSON-RPC Implementations](https://awesome-repositories.com/f/networking-communication/json-rpc-implementations.md) — Implements JSON-RPC for transmitting requests and notifications to servers using stateless HTTP transport. ([source](https://docs.langchain.com/langsmith/agent-server-api/mcp/mcp-post.md))
- [Agent-to-Agent JSON-RPC](https://awesome-repositories.com/f/networking-communication/json-rpc-implementations/agent-to-agent-json-rpc.md) — Facilitates the exchange of messages and task state tracking between AI agents using JSON-RPC over HTTP. ([source](https://docs.langchain.com/langsmith/agent-server-api/a2a/a2a-json-rpc.md))
- [Thread State History](https://awesome-repositories.com/f/networking-communication/messaging-channel-management/history-retrieval/thread-state-history.md) — Fetches past states for specific threads to reconstruct previous agent states using metadata filters. ([source](https://docs.langchain.com/langsmith/agent-server-api/threads/get-thread-history.md))
- [Agent Thread State Persistence](https://awesome-repositories.com/f/networking-communication/messaging-channel-management/history-retrieval/thread-state-history/agent-thread-state-persistence.md) — Persists conversation history and checkpoints to allow agents to maintain context across sessions.
- [Real-time Event Streams](https://awesome-repositories.com/f/networking-communication/real-time-event-streams.md) — Sends structured command and event data over SSE or WebSockets to provide real-time activity updates. ([source](https://docs.langchain.com/langsmith/agent-server-api/system/api-documentation.md))

### Software Engineering & Architecture

- [Tool Exposure Interfaces](https://awesome-repositories.com/f/software-engineering-architecture/tool-exposure-interfaces.md) — Wraps internal agent functions and graphs to expose them as compliant tools for external protocol clients.
- [Long-Running Task Output Streams](https://awesome-repositories.com/f/software-engineering-architecture/client-server-architecture/long-running-task-protocols/long-running-task-output-streams.md) — Streams real-time output from active agent runs with support for resuming from specific event IDs. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/join-run-stream.md))
- [Stateful Execution Resumption](https://awesome-repositories.com/f/software-engineering-architecture/concurrency-schedulers/deterministic-runners/execution-replay/stateful-execution-resumption.md) — Enables restoring execution to prior states to investigate errors or explore alternative agent execution paths. ([source](https://docs.langchain.com/langsmith/core-capabilities.md))
- [Data Validation Schemas](https://awesome-repositories.com/f/software-engineering-architecture/data-validation-schemas.md) — Enforces consistent input and output formats using predefined JSON schemas for reliable inter-agent communication.
- [Stateless Response Streaming](https://awesome-repositories.com/f/software-engineering-architecture/execution-graphs/stateless-execution/stateless-response-streaming.md) — Provides incremental delivery of responses via server-sent events for stateless graph or assistant executions. ([source](https://docs.langchain.com/langsmith/agent-server-api/stateless-runs/create-run-stream-output.md))
- [Asynchronous Agent Job Execution](https://awesome-repositories.com/f/software-engineering-architecture/system-internals/centralization-patterns/workflow-execution-managers/managed-ai-job-execution/asynchronous-agent-job-execution.md) — Provides the ability to trigger and manage long-running agent tasks asynchronously without blocking the caller. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/create-background-run.md))

### Part of an Awesome List

- [Agent Execution Traces](https://awesome-repositories.com/f/awesome-lists/devops/observability-and-tracing/agent-execution-traces.md) — Transforms execution traces into structured dataset entries for monitoring and replay debugging. ([source](https://docs.langchain.com/langsmith/dataset-json-types.md))
- [Agent Frameworks](https://awesome-repositories.com/f/awesome-lists/ai/agent-frameworks.md) — Bridges Anthropic MCP tools with agent workflows.
- [Developer Tools](https://awesome-repositories.com/f/awesome-lists/ai/developer-tools.md) — Adapters to make Model Context Protocol tools compatible with AI agents.
- [LangGraph Extensions](https://awesome-repositories.com/f/awesome-lists/ai/langgraph-extensions.md) — Compatibility layer for using MCP tools within LangChain environments.

### Data & Databases

- [Schema Enforcement Tools](https://awesome-repositories.com/f/data-databases/schema-enforcement-tools.md) — Validates input and output data formats using JSON schemas to ensure processing consistency. ([source](https://docs.langchain.com/langsmith/dataset-json-types.md))
- [Assistant Subgraph Retrieval](https://awesome-repositories.com/f/data-databases/subgraph-extractions/assistant-subgraph-retrieval.md) — Enables fetching of subgraphs associated with an assistant, filtered by namespace with recursive options. ([source](https://docs.langchain.com/langsmith/agent-server-api/assistants/get-assistant-subgraphs.md))

### Security & Cryptography

- [OAuth Provider Integrations](https://awesome-repositories.com/f/security-cryptography/oauth-authentication/oauth-provider-integrations.md) — Enables registration and configuration of external OAuth providers like Google and GitHub for secure access. ([source](https://docs.langchain.com/langsmith/agent-auth.md))
- [OAuth Token Validators](https://awesome-repositories.com/f/security-cryptography/security/policies/token-validation/oauth-token-validators.md) — Verifies the existence and validity of active OAuth tokens for specific providers across associated agents. ([source](https://docs.langchain.com/api-reference/auth-service-v2/check-oauth-token-exists.md))

### System Administration & Monitoring

- [AI Agent Execution Monitors](https://awesome-repositories.com/f/system-administration-monitoring/agent-performance-monitoring/ai-agent-execution-monitors.md) — Provides real-time monitoring of active agent executions on specific threads until completion. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/join-run.md))
- [Scheduled Task Deletion](https://awesome-repositories.com/f/system-administration-monitoring/background-task-scheduling/scheduled-task-deletion.md) — Provides a mechanism to remove periodic tasks by identifier to stop automated background recurrence. ([source](https://docs.langchain.com/langsmith/agent-server-api/crons/delete-cron.md))
- [Run Metadata Retrieval](https://awesome-repositories.com/f/system-administration-monitoring/logging/metrics-retrieval/metric-detail-retrieval/run-metadata-retrieval.md) — Enables retrieval of current status, metadata, and execution history for specific agent invocations. ([source](https://docs.langchain.com/langsmith/agent-server-api/thread-runs/get-run.md))

### Web Development

- [Resumable Event Streams](https://awesome-repositories.com/f/web-development/event-streaming/resumable-event-streams.md) — Implements resumable event streams over SSE for agent threads, including sequence-based reconnection to replay missed data. ([source](https://docs.langchain.com/langsmith/agent-server-api/streaming/protocol-v2-event-stream-sse.md))