Skip to content
EN
English 简体中文 soon 日本語 soon

Augment Code

Repository-wide context for coding

Visit official site

What Augment Code is

Augment Code is an AI coding platform built around a context engine. It reads the entire repository and its dependencies rather than only the open file, and provides chat, completion and step-by-step modification inside VS Code, JetBrains and Vim.

Agents can run locally or in the cloud to plan tasks, make changes and open pull requests, and Model Context Protocol connections reach tools such as GitHub, Jira, Notion and Slack.

What you can do with it

  • Ask questions grounded in the whole codebase
  • Make cross-file changes and dependency upgrades
  • Let an agent implement a task and open a PR
  • Connect code work to issue trackers and chat

Who it is for

  • Enterprise engineering teams
  • Maintainers of large codebases
  • Organisations with privacy and compliance requirements

What to watch out for

  • Repository-wide access means deciding what code and credentials the platform can reach
  • Agent-authored pull requests still need the same review, tests and CI as human ones
  • Broad context helps, but it does not guarantee correct refactors
  • Confirm the current plan limits and how deployment and data handling work

Pros & cons

✓ What we like

  • Context across the whole repository
  • Agents that can finish a task and open a PR
  • Works with common development tools

! What to watch out for

  • Access scope needs governance
  • PRs still need review
  • Plan and data handling to confirm

FAQ

What makes it different from a chat assistant?

The context engine reads the whole repository and dependencies rather than one file.

Can agents submit changes?

Yes. Local and cloud agents can plan tasks, execute changes and open pull requests.

Does it connect to other tools?

Yes. GitHub, Jira, Notion and Slack are mentioned through MCP connections.

Last reviewed: 2026-09-17

More AI coding tools tools

View all →

How we review