Augment Code

Claude Sonnet 4.5 from Anthropic is now the default model for Augment Code. We’re rolling it out to all customers over the next 24 hours, where it will be available alongside Sonnet 4 (for a limited time) and GPT-5 in the model picker.

Our early take on Sonnet 4.5 as compared to Sonnet 4? ✅ 34% fewer tool calls on average ✅ ~26% faster overall task completion time The result: the same accuracy, delivered with more speed and flow.

The Auggie CLI is now available for everyone. Auggie brings Augment’s industry-leading context engine to every part of your stack—terminal, CI, and beyond. Getting started is simple: npm install -g @augmentcode/auggie www.augmentcode.com/product/CLI

Most people think devs use AI to write code from scratch. But we analyzed 81 million developer chats — and that’s not what’s happening. Here's what we found 👇

Most Agent failures aren’t about bad models. They’re about bad prompts. Here’s how to write prompts that actually work—based on thousands of real dev-Agent interactions 👇🧵

Ask for a plan before action. “I need to expose time zone settings. First, suggest a plan—don’t write code yet.” This gives you control. And gives the Agent a checkpoint to align.

Most Agent failures aren’t about bad models. They’re about bad prompts. Here’s how to write prompts that actually work—based on thousands of real dev-Agent interactions 👇🧵

Point the Agent to the right files. ❌ “Add JSON parser to chat backend” ✅ “Add JSON parser in LLMOutputParsing (services/ folder). It’ll be used to extract structured output from chat completions.” Precision = performance.

Most Agent failures aren’t about bad models. They’re about bad prompts. Here’s how to write prompts that actually work—based on thousands of real dev-Agent interactions 👇🧵

Include why, not just what. ❌ “Use events instead of direct method calls” ✅ “Reviewers flagged tight coupling in SettingsWebviewPanel.statusUpdate(). Let’s refactor to events to improve modularity.” Reasoning aligns the Agent with your intent.

Most Agent failures aren’t about bad models. They’re about bad prompts. Here’s how to write prompts that actually work—based on thousands of real dev-Agent interactions 👇🧵

With Augment Rules, your software agent can build just like your team does.

Already using .augment-guidelines.md? No changes required—your setup remains supported. But Augment Rules offers even greater flexibility and control.

With Augment Rules, your software agent can build just like your team does.

Get started in seconds: 🧠 Smart Rule Selection: Agent Requested mode finds what’s relevant for each task 🚀 Seamless Migration: Import rules from other tools, or use your existing Augment guidelines 🧩 Flexible Organization: Use any file name or structure to match your workflow

With Augment Rules, your software agent can build just like your team does.

Prompt. Wait. It messes up. Start over? Not if you build a feedback loop. Here’s how to make Agents actually useful 👇

Failure is feedback. It tells you: – What the Agent misunderstood – What you didn’t explain – What to clarify next Don’t bail—refine.

Prompt. Wait. It messes up. Start over? Not if you build a feedback loop. Here’s how to make Agents actually useful 👇

The best Agent workflows look like test-driven development: Write → run → fix → rerun. You’re not aiming for a perfect prompt—you're building momentum.

Most Agent failures aren’t model problems. They’re context problems. If you give vague or incomplete info, the Agent will fill in the blanks—and usually get them wrong. Here’s how to write high-context prompts that actually work 👇

TL;DR: 🛠️ Tools help 📁 File paths help 🧠 Rationale helps 📌 Examples help Agents don’t need perfect prompts. They need complete ones.

Most Agent failures aren’t model problems. They’re context problems. If you give vague or incomplete info, the Agent will fill in the blanks—and usually get them wrong. Here’s how to write high-context prompts that actually work 👇

A high-context prompt answers the silent questions: - Where should I look? - What else is relevant? - Are there examples I can follow? - What’s the user really trying to do? Agents don’t ask. So you have to pre-answer.

Most Agent failures aren’t model problems. They’re context problems. If you give vague or incomplete info, the Agent will fill in the blanks—and usually get them wrong. Here’s how to write high-context prompts that actually work 👇

Prompting isn’t about being clever. It’s about giving the Agent everything it needs to not guess. That means: - Clear goals - Relevant files - Helpful examples - Precise constraints Agents don’t hallucinate randomly—they hallucinate from missing context.

Agents respond to feedback — just not how you think. The fastest way to improve output quality over time? ✅ Reinforce what works ✅ Redirect what doesn’t ✅ Keep the feedback loop tight Here’s how smart feedback makes agents smarter 👇

Why this works: - Encouragement makes prompts more grounded - Corrections become more specific - The model anchors on what not to change It’s the difference between rewriting from scratch… vs. iterating like a teammate.

Agents respond to feedback — just not how you think. The fastest way to improve output quality over time? ✅ Reinforce what works ✅ Redirect what doesn’t ✅ Keep the feedback loop tight Here’s how smart feedback makes agents smarter 👇

When something works, say so: “Nice — this covers the main use case. Let’s just handle the null edge case too.” This encourages continuation, not correction. It keeps momentum high — and quality stable.

Agents respond to feedback — just not how you think. The fastest way to improve output quality over time? ✅ Reinforce what works ✅ Redirect what doesn’t ✅ Keep the feedback loop tight Here’s how smart feedback makes agents smarter 👇