What happened: How We Made Our Docs Test Themselves

LangChain used Deep Agents CLI to automate migrating inline code samples in documentation into testable, CI-integrated code files.

Why it matters: How We Made Our Docs Test Themselves

Documentation teams need code samples to stay current as APIs evolve and want to catch broken examples without manual maintenance work.

Watch out: How We Made Our Docs Test Themselves

The approach requires upfront setup of skills and CI workflows; it solves the migration problem but doesn't address keeping samples updated as products change.

In the news

How We Made Our Docs Test Themselves

LangChain · Published October 13, 2026 · 3 min read

In 30 seconds

What happened: LangChain used Deep Agents CLI to automate migrating inline code samples in documentation into testable, CI-integrated code files.
Why it matters: Documentation teams need code samples to stay current as APIs evolve and want to catch broken examples without manual maintenance work.
Watch out: The approach requires upfront setup of skills and CI workflows; it solves the migration problem but doesn't address keeping samples updated as products change.

Read the full article

agent

The patterns behind this

Deep Research Agent
Structure-Aware Codebase Retrieval (Repo Map)
RL from Verifiable Rewards (RLVR)

Each one covers how the technique works, when it earns its cost, and where it breaks.

The Agent Architect

One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.

Weekly email, one-click unsubscribe. We only use your address to send the briefing.