Mason Daugherty
e5094037f9
New-contributor onboarding currently requires stitching knowledge together from `AGENTS.md`, per-package `Makefile`s, and external docs, and neither the cross-layer request flow nor the `libs/code` process model is mapped anywhere in the repo. This adds: - Root `ARCHITECTURE.md` (the three-layer `deepagents` → `create_agent` → LangGraph stack and request flow) and `DEVELOPMENT.md` (a single bootstrap + command reference), cross-linked from `README.md` and `libs/README.md`. - `libs/code/ARCHITECTURE.md` (process model, interactive/headless request lifecycles, module map, and a "where do I change X" cheat sheet), linked from `DEV.md` and `AGENTS.md`. Made by [Open SWE](https://openswe.vercel.app) --------- Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
deploy-coding-agent
An autonomous coding agent deployed with deepagents deploy. Given a task description, it plans, implements, tests, and commits changes inside a LangSmith sandbox with full shell access.
Prerequisites
| Variable | Description |
|---|---|
ANTHROPIC_API_KEY |
Claude model access |
LANGSMITH_API_KEY |
Required for deploy and the LangSmith sandbox |
Copy .env.example to .env and fill in both keys.
Deploy
deepagents deploy
The agent is deployed using the config in agent.json.
What to try
Once deployed, open the agent in LangSmith and send it tasks like:
"Add a function that reverses a string and write a test for it""Find all TODO comments in the repo and create a summary""Refactor the main module to use dataclasses"
The agent follows a Plan → Implement → Review → Deliver workflow defined in AGENTS.md.
Structure
deploy-coding-agent/
├── AGENTS.md # Agent instructions and workflow
├── agent.json # Deploy config (name, model)
└── skills/
├── code-review/ # Code review skill with lint helper
├── coding-prefs/ # Coding style preferences
└── planning/ # Task planning skill
MCP servers: This example previously used
mcp.jsonto wire in the LangChain docs MCP server. MCP servers are now workspace-level resources. Register them once withdeepagents mcp-servers add --url <url>and reference them in atools.jsonfile.
Query via SDK
from langgraph_sdk import get_client
client = get_client(url="https://<your-deployment-url>")
thread = await client.threads.create()
async for chunk in client.runs.stream(
thread["thread_id"], "agent",
input={"messages": [{"role": "user", "content": "Add a hello_world function and test it"}]},
stream_mode="messages",
):
print(chunk.data, end="", flush=True)
Find your deployment URL in LangSmith under Deployments. See the LangGraph SDK docs for more.
Resources
- deepagents deploy docs
- LangSmith sandbox docs
- LangChain Academy — Comprehensive, free courses on LangChain libraries and products, made by the LangChain team.
- Code of Conduct — community guidelines and standards