deepagents is a Doramagic Project Pack for Structured LLM app starting paths, RAG/tool-calling checks, migration reminders, permission boundaries, and acceptance checks. It links the upstream repository, source-backed manual, prompt preview, AI context, and risk boundaries so users can evaluate the project before installation.

What should I verify first?

Verify install, import, and one minimal task path in a temporary Python environment before using a primary project.. Keep credentials out of the first run, record expected output, define cleanup, and use public sample data before real workloads.

What is the main risk?

May increase setup, validation, or first-run risk for the user.. Use upstream documentation and a sandbox validation path before adoption.

LLM Application Framework · Public

deepagents

LLM application framework for checking model, prompt, tool, retrieval, and chain integration boundaries.

LLM appsRAGTool callingPythonFramework migration

Best fitDevelopers building Python LLM apps, RAG workflows, tool calling, or agent prototypes that need a shared abstraction layer.

Check whether this project matches your task before installing it.

What it can doStructured LLM app starting paths, RAG/tool-calling checks, migration reminders, permission boundaries, and acceptance checks

Review the portable capability path.

Before continuingVerify in a sandbox

Do not treat a preview pack as a proven local install.

GitHub snapshot23k stars

3.3k forks · 119 contributors

Doramagic.ai Last verification date: 2026-06-01 Verification method: source evidence, semantic profile, public page gate, and static build acceptance.

Official first step Read manual preview Source repository

Publication status · 2026-06-01

What is deepagents?

deepagents is an LLM application framework for model, prompt, tool, retrieval, chain, and callback interfaces.
Best fit: Developers building Python LLM apps, RAG workflows, tool calling, or agent prototypes that need a shared abstraction layer.
Not for: Not for one-off model API calls, simple prompting, or stateful agent orchestration work that should first compare LangGraph-style options.
Capability added to an AI workflow: Structured LLM app starting paths, RAG/tool-calling checks, migration reminders, permission boundaries, and acceptance checks
First safe verification step: Verify install, import, and one minimal task path in a temporary Python environment before using a primary project.
Verification state: source, Quick Start, and sandbox install checks are recorded as passed.
Top risk: May increase setup, validation, or first-run risk for the user.
Evidence base: https://github.com/langchain-ai/deepagents, https://github.com/langchain-ai/deepagents#readme, Human Manual, Pitfall Log

Quick decision

Use this section to decide whether the project is worth a deeper read.

Best forDevelopers building Python LLM apps, RAG workflows, tool calling, or agent prototypes that need a shared abstraction layer.

Match the project to your task before installing it.

CapabilityStructured LLM app starting paths, RAG/tool-calling checks, migration reminders, permission boundaries, and acceptance checks

LLM application framework for checking model, prompt, tool, retrieval, and chain integration boundaries.

Repositorylangchain-ai/deepagents

23k stars · 3.3k forks

What it can do

Translate the upstream project into concrete capabilities the user can judge before installing.

Deep Agents Overview

Related topics: Quickstart Guide, Agent Graph Architecture

Source: https://github.com/langchain-ai/deepagents / Human Manual

Quickstart Guide

Related topics: Deep Agents Overview, Agent Graph Architecture, Skills System

Source: https://github.com/langchain-ai/deepagents / Human Manual

Agent Graph Architecture

Related topics: Middleware System, Filesystem and Sandbox Backend, Subagents System

Source: https://github.com/langchain-ai/deepagents / Human Manual

Middleware System

Related topics: Agent Graph Architecture, Filesystem and Sandbox Backend, Memory Middleware, Skills System, Context Summarization

Source: https://github.com/langchain-ai/deepagents / Human Manual

Filesystem and Sandbox Backend

Related topics: Middleware System, Memory Middleware

Source: https://github.com/langchain-ai/deepagents / Human Manual

Sources: https://github.com/langchain-ai/deepagents, Human Manual, Project Pack evidence, and downstream validation signals.

Community Discussion Evidence

Project-level external discussion stays visible on the detail page, not only inside the manual.

Stars23k stars

Forks3.3k forks

Contributors119 contributors

Licenseunknown

Community Discussion Evidence

9 source-linked items

Review these external discussions before using deepagents with real data or production workflows. They are review inputs, not standalone proof that the project is production-ready.

01
Add `RemoteBackend` as a backend alternative
github / github_issue
02
feat(sdk): add regex support to the `grep` tool
github / github_issue
03
Filter by agent in `/threads`
github / github_issue
04
`FilesystemMiddleware`'s `_handle_read_result` ignores `read_result.file
github / github_issue
05
`.deepagentsignore`
github / github_issue
06
`read_file` pagination skips lines after wrapping due to double limit ap
github / github_issue
07
Summarization middleware image loss
github / github_issue
08
MemoryMiddleware: add_cache_control=True (0.6) puts incorrect cache_cont
github / github_issue
09
Capability evidence risk requires verification
GitHub / issue

How to start

Only source-backed commands are shown here. Verify them in an isolated environment first.

Try the prompt first

Test the workflow without installing the upstream project.

preview

Read the Human Manual

Understand inputs, outputs, limits, and failure modes.

manual

Take context to your AI host

Use the compiled assets in your preferred AI environment.

context

Run sandbox verification

Confirm install commands and rollback before using a primary environment.

verify

pip install deepagents

Official start command · https://github.com/langchain-ai/deepagents#readme · verified: yes

Human Manual

The English page must expose the real manual, not a short placeholder.

8+ sections · Human Manual

deepagents Manual

Related topics: Quickstart Guide, Agent Graph Architecture

Open the full manual

https://github.com/langchain-ai/deepagents Project Manual
Table of Contents
Deep Agents Overview
Related Pages
Architecture Overview
Relationship to LangChain Ecosystem
Core Components
Agent Creation

Deep Agents Overview

Related topics: Quickstart Guide, Agent Graph Architecture

Source: https://github.com/langchain-ai/deepagents / Human Manual

Quickstart Guide

Related topics: Deep Agents Overview, Agent Graph Architecture, Skills System

Source: https://github.com/langchain-ai/deepagents / Human Manual

Agent Graph Architecture

Related topics: Middleware System, Filesystem and Sandbox Backend, Subagents System

Source: https://github.com/langchain-ai/deepagents / Human Manual

Middleware System

Related topics: Agent Graph Architecture, Filesystem and Sandbox Backend, Memory Middleware, Skills System, Context Summarization

Source: https://github.com/langchain-ai/deepagents / Human Manual

Filesystem and Sandbox Backend

Related topics: Middleware System, Memory Middleware

Source: https://github.com/langchain-ai/deepagents / Human Manual

AI Context Pack and portable assets

After deciding to continue, take the project context into your own AI host.

Complete pack plus user-owned assets

These files are planning and verification assets for Claude Code, Codex, Gemini, Cursor, ChatGPT, and other AI hosts.

Download complete pack Read Human Manual

BundleComplete Project Pack AssetAI Context Pack AssetBoundary & Risk Card AssetHuman Manual AssetPitfall Log AssetPrompt Preview AssetQuick Start EvidenceREPO_INSPECTION.json

Preflight checks

Treat this page as a planning asset, not proof that your local environment is ready.

The manual is generated from source-linked project files and Doramagic validation signals.
Community evidence warnings stay visible instead of being converted into marketing claims.
This English page is indexable because the locale quality gate passed and explicit English index approval is enabled.
Use the upstream repository as the final authority for installation commands, license, and version-specific behavior.

Pitfall Log and verification risks

Doramagic surfaces high-risk items before users treat a candidate capability as verified.

high

Configuration risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Installation risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Configuration risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Capability evidence risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Maintenance risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Security or permission risk requires verification

May increase setup, validation, or first-run risk for the user.

medium

Security or permission risk requires verification

May increase setup, validation, or first-run risk for the user.

low

Maintenance risk requires verification

May increase setup, validation, or first-run risk for the user.