Google's Agents CLI gets an Extension System to support LangChain and other frameworks beyond ADK
In short: Google has added an Extension System to its Agents CLI (from v1.5.0, with a first-party LangChain extension in v1.8.0) that lets developers override or add CLI commands without forking the tool. Previously, Agents CLI's deployment, evaluation and observability pipeline was tightly coupled to Google's own Agent Development Kit (ADK); now teams can scaffold, run, deploy and evaluate LangChain, LangGraph or Deep Agents projects through the same command surface. Extensions are distributed as directories in Git repositories, pinned to a commit SHA in a lockfile, and can be installed at project or user scope. The system also lets enterprise platform teams wrap commands like deploy with custom policies, such as compliance scans or SSO gates. The feature is marked experimental in v1.8.0.
This summary was generated automatically by AI from Google Developers's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.
What changed?
- 1Extension System added to Agents CLI starting v1.5.0
- 2First-party LangChain extension shipped in v1.8.0, scaffolding LangChain/LangGraph/Deep Agents projects with zero google-adk dependency
- 3Extensions are Git-repo directories declared via an agents-cli-extension.yaml manifest, either overriding built-in commands or adding new ones
- 4Extensions resolve to a pinned commit SHA via a lockfile, re-materialized identically on every checkout
- 5Project scope (committed to repo) and user/global scope (per machine) supported, with project scope taking priority on conflicts
- 6Enterprise teams can wrap commands like deploy with custom checks (e.g. SBOM scans, compliance gates)
- 7Marked experimental in v1.8.0; manifest schema may still change
Why it matters
Agent development teams are no longer forced to adopt a single framework just to get Google's deployment, evaluation and observability tooling for Cloud Run; this reduces the cost of switching or mixing frameworks, and lets platform teams enforce governance policies consistently across projects regardless of the agent framework used.
What it means for AI agents and contact centers
For teams building voice agents with tool/function calling and MCP, this makes it easier to prototype in LangChain or LangGraph while still deploying to Cloud Run with the same observability and eval pipeline used for ADK agents — useful when comparing frameworks for call-routing or contact-center automation agents without committing to a single stack upfront.
🧪 Worth Testing
Worth testing if evaluating agent frameworks for voice/contact-center use cases, since it removes the overhead of separately wiring deployment and evaluation pipelines for each framework choice.
Sources
- Google DevelopersOfficialPrimary sourceOriginal article →„Powering the Agent Development Lifecycle with any framework, from ADK to LangChain and beyond“21 Oct 2026, 03:00
- Published by source
- 21 Oct 2026, 03:00
- Found by our system
- 21 Oct 2026, 20:51
- Summary generated
- 21 Oct 2026, 20:55
This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy