deepagents
Agent harness built with LangChain and LangGraph. Equipped with a planning tool, a filesystem backend, and the ability to spawn subagents - well-equipped to handle complex agentic tasks.
Learn to add custom middleware to the Deep Agent execution stack by subclassing AgentMiddleware and overriding hooks. Enhance your agent's functionality today.
Security Model and Isolation Strategy for Execute Command Sandboxing in DeepAgentsExplore the three-layer security model for DeepAgents execute command sandboxing, featuring container isolation, secure payloads, and configurable timeouts. Learn how it prevents command injection.
How the Harbor Framework Facilitates Evaluation and Benchmarking of Agent PerformanceDiscover how the Harbor framework simplifies agent evaluation and benchmarking. It runs Deep Agents in isolated sandboxes, executes benchmarks, and records traces for deterministic scoring and analysis.
How to Migrate Agent Logic from the DeepAgents SDK to the CLI for Interactive DevelopmentMigrate agent logic from DeepAgents SDK to CLI using create_cli_agent for interactive development. Enjoy automatic middleware setup, interactive prompts, and Textual UI configuration.
TodoListMiddleware vs Other Planning Tools in DeepAgents: Key Functional DifferencesDiscover the functional differences between TodoListMiddleware and other planning tools in DeepAgents. Learn how TodoListMiddleware's dedicated checklist API contrasts with auxiliary workflows in SubAgentMiddleware and HumanInT...
DeepAgents Prompt Caching for Anthropic: Reducing LLM Latency with Automatic MiddlewareReduce Anthropic LLM latency with DeepAgents prompt caching. Automatically cache Claude API responses to eliminate redundant calls and speed up your agents without manual setup.
How to Configure the interrupt_on Parameter in DeepAgents for Human-in-the-Loop WorkflowsLearn to configure the interrupt_on parameter in DeepAgents for custom human approval workflows. This guide explains how to set up middleware for efficient human-in-the-loop processes.
Deep Agents CLI Configuration Options for Remote Sandboxing Environments Like DaytonaExplore Deep Agents CLI configuration options for remote sandboxing with Daytona. Learn to use --sandbox --sandbox-id and --sandbox-setup flags for seamless integration.
How the DeepAgents CLI Manages Persistent Agent Sessions and Stateful MemoryDiscover how the DeepAgents CLI manages persistent agent sessions and stateful memory by saving LangGraph checkpoints to SQLite and injecting user defined memory files into system prompts.
PatchToolCallsMiddleware: Ensuring Tool Call Consistency in DeepAgentsDiscover PatchToolCallsMiddleware in DeepAgents to ensure tool call consistency and message history integrity in your agent pipelines automatically. Learn how it handles missing tool responses.
How to Integrate a Custom Backend Implementation Using the BackendProtocol in DeepAgentsLearn to integrate a custom backend with BackendProtocol in DeepAgents. Implement the interface, override file operations, and pass it to DeepAgent for seamless integration.
How DeepAgents Implements Tool Retries and Error Recovery During Agent ExecutionDeepAgents uses a tiered detect classify recover strategy with middleware for robust tool retries and error recovery without breaking agent execution. Learn how.
Have a question about this repo?
These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:
curl -s "https://instagit.com/install.md" 
Maintain an open-source project? Get it listed too →