llms.txt
# Augment Code
> Augment Code is an AI-powered coding assistant that helps developers write better code faster with deep IDE integration and context-aware suggestions.
Augment Code provides:
- AI-powered code completion and generation with industry-leading accuracy
- Deep integration with popular IDEs (VS Code, JetBrains, and more)
- Context-aware suggestions using our world-class context engine
- Support for multiple programming languages and frameworks
- Model Context Protocol (MCP) server integrations
## When to use this
- Use the Augment Docs MCP server to answer current questions about Augment setup, Cosmos, Auggie CLI, IDE integrations, configuration, and troubleshooting. Connect over Streamable HTTP at https://docs.augmentcode.com/mcp.
- Use the Analytics API for organization-level usage, active-user, credit, cost, and budget reporting. Read /openapi.json first, authenticate with a scoped OAuth token or service-account bearer token, and request write access only when changing budget overrides.
- Use Auggie CLI for repository-scoped implementation, codebase exploration, test execution, and CI automation where deep code context matters. Start with the CLI documentation at https://docs.augmentcode.com/cli/overview.
- Use Cosmos for long-running or multi-step software-delivery work that needs durable environments, experts, integrations, scheduled automations, or coordination across repositories and tools. Start with https://docs.augmentcode.com/cosmos/getting-started.
## Guides and Tutorials
- [CodeRabbit Pricing 2026: What It Costs (and When an Alternative Makes Sense)](https://www.augmentcode.com/guides/coderabbit-pricing-costs-limits)
- [From Interactive Sessions to Agent-Run Pipelines: Where the Software Factory Gets Cheaper](https://www.augmentcode.com/guides/managed-agents-vs-interactive-sessions)
- [Budgeting the Agent Fleet: Tokens, Compute, and Headcount](https://www.augmentcode.com/guides/budgeting-ai-agents-vs-headcount)
- [Human in the Loop vs Human on the Loop: Where the Reviewer Sits in an Agent-Run SDLC](https://www.augmentcode.com/guides/human-in-the-loop-vs-human-on-the-loop)
- [The Engineering Manager's Job in a Software Factory](https://www.augmentcode.com/guides/engineering-manager-ai-software-factory)
- [The Dark Factory: What Fully Autonomous Software Delivery Actually Requires](https://www.augmentcode.com/guides/dark-factory-autonomous-software-delivery)
- [Can AI Agents Safely Approve Pull Requests? Guardrails for Autonomous Code Review](https://www.augmentcode.com/guides/ai-agents-pull-request-approval-guardrails)
- [Cursor Bugbot for Code Review: Capabilities and Limits](https://www.augmentcode.com/guides/cursor-bugbot-code-review-capabilities-limits)
- [Standardizing AI-Assisted Development: Variance Is the Real Constraint](https://www.augmentcode.com/guides/standardizing-ai-assisted-development-variance)
- [GitHub Spec Kit: A Practical Guide and Where It Fits Your SDD Workflow](https://www.augmentcode.com/guides/github-spec-kit-practical-guide-sdd-workflow)
## MCP Server Integrations
- [context7](https://www.augmentcode.com/mcp/context7): Context7 MCP Server – serves fresh, version-specific code documentation to Model-Context-Protocol (MCP) clients such as Cursor, Claude Code or Opencode.
- [prisma](https://www.augmentcode.com/mcp/prisma): Next-generation ORM for Node.js & TypeScript supporting PostgreSQL, MySQL, MariaDB, SQL Server, SQLite, MongoDB and CockroachDB.
- [mindsdb](https://www.augmentcode.com/mcp/mindsdb): AI’s query-engine & MCP server for building agents that answer questions over large-scale, federated data sources.
- [composio](https://www.augmentcode.com/mcp/composio): Composio equips AI agents & LLMs with 100+ ready-made integrations via function calling (official TypeScript & Python SDKs + Rube MCP server).
- [Taskmaster MCP Server](https://www.augmentcode.com/mcp/taskmaster-mcp-server): Taskmaster is an AI-powered task-management system for AI-driven development that runs as an MCP server (for editors like Cursor, Windsurf, VS Code, Amazon Q CLI) and also as a CLI. Install via MCP using `npx -y task-master-ai` (or `npm i -g task-master-ai`) and provide at least one model/provider credential (e.g., Anthropic/OpenAI/Google/Perplexity/xAI/OpenRouter/etc.) or use Claude Code/Codex CLI. Supports selective MCP tool loading via `TASK_MASTER_TOOLS` (all/standard/core or custom list) and provides tools for project initialization, PRD parsing, task planning/expansion, task status updates, and research workflows.
- [FastMCP](https://www.augmentcode.com/mcp/fastmcp): 🚀 The fast, Pythonic way to build MCP (Model Context Protocol) servers and clients. FastMCP is a Python framework that lets you expose Python functions as MCP tools (with automatic schema/validation/docs), connect to MCP servers as a client (URL-based connection with transport negotiation/auth/protocol lifecycle), and build interactive "Apps" (UIs rendered in-conversation). Installation (recommended): `uv pip install fastmcp`. Docs: https://gofastmcp.com. Maintained by Prefect (PrefectHQ).
- [github-mcp-server](https://www.augmentcode.com/mcp/github-mcp-server): GitHub’s official Model Context Protocol (MCP) server that exposes rich, tool-oriented access to GitHub’s REST & GraphQL APIs. Supports remote (hosted) or local (Docker / native Go binary) deployment and fine-grained toolset configuration.
- [blender-mcp](https://www.augmentcode.com/mcp/blender-mcp): Blender Model Context Protocol (MCP) server that lets Claude AI (and other MCP-aware IDEs such as Cursor, VS Code, and Claude Desktop) drive a live Blender session for prompt-assisted 3D modelling, scene inspection, and code execution inside Blender.
- [screenpipe](https://www.augmentcode.com/mcp/screenpipe): AI app-store platform that continuously records a user’s desktop (screen + mic) locally, indexes it, and exposes an API so developers can build context-aware AI desktop apps (“pipes”) in Next.js, publish them and monetise through the built-in store.
- [playwright-mcp](https://www.augmentcode.com/mcp/playwright-mcp): Playwright MCP server
## Blog Posts
- [Augment Code's next chapter](https://www.augmentcode.com/blog/augment-code-next-chapter)
- [Beyond AI Coding Agents: How We Built Augment's Software Factory](https://www.augmentcode.com/blog/beyond-ai-coding-agents-how-we-built-augments-software-factory)
- [We think the FDE should be part of the product, so we built one in](https://www.augmentcode.com/blog/we-think-the-fde-should-be-part-of-the-product-so-we-built-one-in)
- [How a two-person engineering team handled 6× more product feedback and kept shipping](https://www.augmentcode.com/blog/feedback-triage-with-cosmos)
- [How we rebuilt the Auggie CLI harness to make tasks 53% cheaper than Claude Code](https://www.augmentcode.com/blog/auggie-cli-harness-rebuild-53-percent-cheaper)
- [The software factory needs a faster review loop: further optimizing the path from PR to merge](https://www.augmentcode.com/blog/optimizing-pr-to-merge-loop)
- [Using the Cosmos software factory to better understand our business](https://www.augmentcode.com/blog/cosmos-software-factory-business-visibility)
- [The best value for every token: GPT-5.6 Sol is now our default in Cosmos](https://www.augmentcode.com/blog/eight-models-in-eight-weeks-gpt-5-6-sol-is-now-our-default)
- [What is loop engineering, and how are leading software engineering teams using it?](https://www.augmentcode.com/blog/what-is-loop-engineering-and-how-are-leading-software-engineering-teams-using-it)
- [The bottleneck moved to verification. So we automated that too.](https://www.augmentcode.com/blog/the-bottleneck-moved-to-verification-so-we-automated-that-too)
## Product Updates
- [VSCode Extension 0.904.1 Release Notes](https://www.augmentcode.com/changelog/vscode-0-904-1-stable-release-notes)
- [Cosmos Week 38 Release Notes](https://www.augmentcode.com/changelog/cosmos-week-38-release-notes)
- [IntelliJ Plugin v0.491.0 Release Notes](https://www.augmentcode.com/changelog/intellij-plugin-v0-491-0-release-notes)
- [VSCode Extension 0.901.1 Release Notes](https://www.augmentcode.com/changelog/vscode-0-901-1-stable-release-notes)
- [Cosmos Week 37 Release Notes](https://www.augmentcode.com/changelog/cosmos-week-37-release-notes)
## Machine Interfaces
- [OpenAPI specification](https://www.augmentcode.com/openapi.json): OpenAPI 3.1 description of Augment Code public APIs
- [MCP discovery manifest](https://www.augmentcode.com/.well-known/mcp): Connection details for the public Augment Docs MCP server
- [MCP server card](https://www.augmentcode.com/.well-known/mcp/server-card.json): Identity and Streamable HTTP transport metadata for the Augment Docs MCP server
## Optional
- [Product Overview](https://www.augmentcode.com/product): Comprehensive overview of Augment Code features and capabilities
- [Pricing](https://www.augmentcode.com/pricing): Pricing plans and feature comparison
- [Security](https://www.augmentcode.com/security): Security practices, compliance, and data protection
- [Careers](https://www.augmentcode.com/careers): Join our team and help shape the future of AI-powered development
- [Legal](https://www.augmentcode.com/legal): Terms of service, privacy policy, and legal information
