RSS

Introducing the Tanagram CLI

Tanagram is now available as a CLI and agent skill — a new way to enforce your team's coding rules directly in Claude Code and other coding agents.

Getting started is simple:

npm install -g @tanagram/cli
tanagram login

That's it. Once installed, Claude Code will automatically check your changes against your team's Tanagram rules. When a rule is violated, Claude sees the feedback and fixes the issue itself.

Under the hood, the CLI adds a Claude Code skill that runs tanagram eval after code changes. Your rules are synced from your Tanagram account and cached locally, so evaluation is fast.

Give it a try and let us know what you think.

Rules, Not Policies

We've renamed 'policies' to 'rules'. The term 'policy' felt a bit too formal; it makes you feel like something that restricts you, rather than something that empowers you. Talking to users, it's also a less familiar term; instead, 'rule' was a term that readily came to mind for everyone.

So, as of today, policies are dead … long live rules.

Suggested Policies Are Generally Available

We previously shipped suggested policies in private beta. They are now generally available for all users as part of our refreshed onboarding flow.

Try them out and let us know relevant the suggestions are for you.

How to Efficiently Manage AI Coding Tools

Here's what nobody tells you about managing AI coding tools: the hardest part isn't getting developers to use them. It's preventing the mess they create when everyone uses them differently.

Suggested Policies Are in Private Beta

For select users, we'll new automatically suggest policies based on your repo history. You can review and accept suggestions, which will turn them into rules that will run on every PR.

If you have access, take a look at the suggestions we've generated for you and let us know how relevant they seem.

Enhancing Code Quality with Automation

Here's what nobody tells you about code quality automation: the tools are easier than the cultural shift. The real challenge isn't picking tools—it's capturing the knowledge your senior engineers carry in their heads and turning it into enforceable guardrails that work at scale.

Improved Policy Editor

We've made it easier to write policies — instead of the plain-text input box we used to have, we've brought in a markdown editor. For users who add a lot of detail to their policies, this should make the authoring and editing experience a lot nicer.

A screenshot of our new policy editor, showing a text box with "Write" and "Preview" tabs. the "Write" tab is selected, showing content that contains markdown syntax

Why Code Review Policy Enforcement Matters More Than Ever in 2025

While the tech industry celebrated AI coding agents hitting 85% adoption across enterprises in 2025, a more critical story was unfolding. The same AI revolution that promised to accelerate development was simultaneously creating an unprecedented risk landscape.