augmentcode-com

Skip to contentProductAgentIntentCode ReviewSlackCLIContext Engine MCPChangelogContext EnginePricingDocsBlogResourcesCustomersCareersContextWikiStatus PageTrust CenterSecurityTalk to SalesSign inInstallBlog: The hardest part about going AI-native isn't just technicalAugment Intent: Orchestrate your agents like a system, not a swarmBlog: The hardest part about going AI-native isn't just technicalAugment Intent: Orchestrate your agents like a system, not a swarmThe Software Agent CompanyBuild software with AI agents that understand your entire codebase.From IDE to CLI to code review, Augment works where you work.Install NowBook a DemoTRUSTED BY ENGINEERING TEAMSMeet our customersCONTEXT ENGINEWhy context quality determines code qualityEvery AI coding tool uses the same models. Augment's Context Engine is the difference. We maintain a live understanding of your entire stack — code, dependencies, architecture, and history.Discover the Context Engine< -500>50Agent UnderperformsHumanAgent OutperformsOverallOverallAggregate performance across functional correctness, style, and context awareness.-13.9-11.8+12.8CorrectnessCorrectnessCode executes as intended, passes tests, and handles edge cases without logical errors.-11.8-9.3+14.8CompletenessCompletenessSolution fully implements the requested feature scope, leaving no placeholders or TODOs.-12.4-12+18.2Code ReuseCode ReuseIntelligently leverages existing project utilities, types, and components to minimize technical debt.-15.8-9.3-4.4Best PracticeBest PracticeCode matches the unique patterns, naming conventions, and architecture of the codebase.-16.4-10.5+12.4Blind study comparing 500 agent-generated pull requests to merged code written by humans on the Elasticsearch repository — 3.6M Java LOC from 2,187 contributorsCursorClaude CodeAugment CodeSWE-Bench Pro LeaderboardAuggie51.80%Cursor50.21%Claude Code49.75%Codex46.47%SWE-AgentScale baseline45.89%44%47%50%53%All agents on Opus 4.5. Better results. It's the context. Read More→INTENTBuild with Intent.A developer workspace where agents are coordinated, specs stay alive, and every workspace is isolated.01Build with a coordinated team of agents02Build with a living spec03Build in an isolated environmentBuild with IntentCoordinator — Intent Product PageIntent Product Pageaugmentcode/dotcomAgentsContextChangesAgents coordinate tasks.+ Create new agentCoordinatorLet's continue with...Code ReviewDesign Review3 backgroundCoordinatorhero-section...Create high-fidelity app mockupBreaking into parallel tasks...Hero Mockup Agent// intent-app-mockup.tsxexport function IntentApp...Mobile View AgentAdding responsive breakpoints...Just nowType a message...SpecNOTES / SpecIntent Product Page SpecOverviewCreate product page for Intent — Augment's macOS desktop app.Page Structure✓Hero Section✓Feature Sections✓SDD Section✓Download CTA◐Redesign Hero◐Mobile View○AnimationsTERMINALAugment CLIAI-powered coding in your terminal. For engineers who prefer the command line. Same Context Engine, same powerful agents, no GUI required.01Full terminal integration02Works with your shell03Same context as IDE agentsTry the CLIauggie-cli··· ██████ █ █ ██████ ██████ █ ██████ █ █ █ █ █ █ █ █ ██████ █ █ █ ███ █ ███ █ █████ █ █ █ █ █ █ █ █ █ █ █ █ ██████ ██████ ██████ █ ██████BY AUGMENT CODEPersonalized prompt suggestions, for you (.•◡•.)◇Remove unnecessary packages: dist, node-sass, sass-loader◇Reorganize dependencies: Moved sass to devDependencies where it belongs◇Added essential dev tools: ESLint, Prettier, TypeScript supportFor automation, use 'auggie --print "your task"'>Try 'how do I log an error?' or type / for commands? to show shortcuts~/Augment/AugmentVS Code — Augment AgentThreadTasks5/22Edits+18-54UAdd rate limiting to the API endpointsAI'll add rate limiting to your API. Let me check the existing middleware setup.Context Engine: Codebaserate limit apiCreating src/middleware/rateLimit.ts+42 -0We like you already.Install Augment nowroutes.tsAutoClaude Opus 4.6IDE AGENTSFrom prompt to pull requestMost AI-generated code needs cleanup. Augment agents are different: our deep contextual understanding of your codebase means the code they write is superior, not slop.Augment is built for pro software teams:01Task lists for complex, multi-step work02Automatic memories across sessions03VS Code & JetBrainsExplore IDE Agentsgithub.com — Pull Request #1842···components/open-source/mdx-components.tsx8+ */9+export const mdxComponents: MDXComponents = {10+ h1: ({ children, ...props }: React.ComponentProps<'h1'>) => {11+ const id = generateHeadingId(String(children));Augment Code Reviewbot3 hours agoUsing String(children) to derive heading IDs can return '[object Object]' when the heading contains inline elements (e.g., code/emphasis), leading to anchor mismatches with the TOC; consider extracting plain text from children (also applies to h2/h3).Was this useful? React with thumbs up or thumbs down+-☆Reply...CODE REVIEWAugment Code ReviewThe only AI code reviewer that thinks like a senior engineer. We benchmarked 7 leading tools on real production codebases—Augment delivered the highest precision and recall by a significant margin. Context-powered reviews catch critical bugs without the noise.01Inline comments in GitHub02Full codebase context via industry-leading Context Engine03One-click fixes in your IDEDiscover Code ReviewGet StartedGive your codebase the agents it deservesInstall Augment to get started. Works with codebases of any size, from side projects to enterprise monorepos.Install AugmentContact Sales --- Skip to contentProductAgentIntentCode ReviewSlackCLIContext Engine MCPChangelogContext EnginePricingDocsBlogResourcesCustomersCareersContextWikiStatus PageTrust CenterSecurityTalk to SalesSign inInstallPRICINGPricing that scales with you & your team.POPULARINDIE$20/monthPerfect to get started, and for developers who use AI a couple of times per week.Install nowIncludes 40,000 creditsContext EngineMCP & Native ToolsSOC 2 type IIAuto top-up credits availableNo AI training allowedCredits can be used for Code ReviewSTANDARD$60/monthper developerFor individuals or small teams that want to ship to production, fast.Install nowEverything in IndieIncludes 130,000 creditsMAX$200/monthper developerDesigned for high-demand teams or businesses with intensive usage needs.Install nowEverything in StandardIncludes 450,000 creditsENTERPRISECustomFor enterprise teams with high volume, security, or support needs.Contact salesCustom user pricingBespoke credit limitSlack integrationVolume based annual discountsSSO, OIDC, & SCIM supportSOC 2 & Security ReportsCMEK & ISO42001 complianceDedicated supportNo AI training allowedCredits can be used for Code ReviewTRUSTED BY ENGINEERS ATMeet our customers001Compare plansINDIE$20STANDARD$60MAX$200ENTERPRISECustomINDIE$20/month40,000 creditsAuto top-up $15/24k creditsUp to 1 userContext EngineCoding AgentIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationCode ReviewPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistSecuritySOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationSTANDARDPOPULAR$60/monthper developer130,000 creditsAuto top-up $15/24k creditsUp to 20 usersContext EngineCoding AgentIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationCode ReviewPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistSecuritySOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationMAX$200/monthper developer450,000 creditsAuto top-up $15/24k creditsUp to 20 usersContext EngineCoding AgentIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationCode ReviewPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistSecuritySOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationENTERPRISECustomCustom creditsCustom top-upUnlimited usersContext EngineCoding AgentIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationCode ReviewPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistSecuritySOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationINDIE$20/monthSTANDARD$60/monthper developerMAX$200/monthper developerENTERPRISECustom40,000 creditsAuto top-up $15/24k creditsUp to 1 userContext Engine130,000 creditsAuto top-up $15/24k creditsUp to 20 usersContext Engine450,000 creditsAuto top-up $15/24k creditsUp to 20 usersContext EngineCustom creditsCustom top-upUnlimited usersContext EngineCoding AgentIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationIntentChat & AgentsMCP & Native ToolsUsage AnalyticsSlack integrationCode ReviewPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistPR Summaries & Inline CommentsAuto and Manual ModePR GuidelinesMCP supportGitHub Multi Org supportAdvanced AnalyticsUser allowlistSecuritySOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationSOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationSOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO IntegrationSOC 2 Type IICMEK & ISO 42001 ComplianceSIEM IntegrationData Residency OptionsGranular Access ControlsComprehensive Audit TrailsEnterprise SSO Integration002Frequently asked questions01What is a credit?A credit is the unit we use to measure usage in Augment Code. For example: A small task that uses 10 tool calls would cost around 300 credits A complex task that requires 60 tool calls would cost around 4,300 credits Learn more about credit-based pricing →02Are credits pooled?Yes. Credits are pooled at the team level. Here is an example for a team of 20 developers on Standard Plan: Total monthly credits: 20 users × 130,000 credits = 2,600,000 credits Typical usage pattern: 3 power users: 330,000 credits each = 990,000 credits 10 regular users: 78,000 credits each = 780,000 credits 7 occasional users: 35,000 credits each = 245,000 credits Total used: 2,015,000 credits (585,000 remaining) 03Do you train on my data on a paid plan?No. All paid plans exclude any AI training on your data as part of our Commercial Terms of Service.04How are you handling Security & Compliance?We take security & compliance seriously. In addition to our rigorous encryption and access controls, we undergo annual SOC 2 audits. Visit our Security page to learn more, or head to our Trust Center to request additional details and reports.05What SLA and support do I get on each plan?All paid subscriptions are covered by the same core uptime and response targets described in our SLA and Support Policy. Support channels differ by plan: Indie & Standard: Community-based support, plus the ability to file tickets via our support portal at support.augmentcode.com. Tickets submitted through the portal are handled under our SLA. Max: Everything in Standard, with email-based support in addition to community channels, handled under the same SLA when you file tickets through our support portal. Enterprise: Full SLA coverage with dedicated support on top of our standard commitments. Free & trial use: Community support only; our SLA does not apply to trials or beta usage. Read the full SLA and Support Policy →06When do my credits expire?Credits top-ups that are not part of your paid plan are valid 12 months after the date of purchase.07What is the difference between Code Review and Code Review Enterprise?Code Review is available on all plans and provides AI-powered code review for GitHub pull requests. Enterprise Code Review includes additional capabilities: Advanced Analytics: Dashboard showing reviews performed and comments addressed User Allowlists: Control which users receive PR reviews MCP Configuration: Connect to ticketing systems (Jira, Linear), docs (Notion), and other tools for richer context Multi-org Support: Install on multiple GitHub organizations Unlimited Seats & Repos: No limits on team size or connected repositories Learn more in the docs.Ready to get started?Feel the difference with AI built for production scale codebases.Install NowTalk to Sales --- Skip to contentProductAgentIntentCode ReviewSlackCLIContext Engine MCPChangelogContext EnginePricingDocsBlogResourcesCustomersCareersContextWikiStatus PageTrust CenterSecurityTalk to SalesSign inInstallThe Context EngineEvery AI uses the same models.Context is the difference.Augment's Context Engine maintains a live understanding of your entire stack—across repos, services, and history. Agents that finish tasks, not just suggest lines.Get StartedHow It Works1M+Files IndexedReal-timeKnowledge Graph100%Your Patterns001Why most agents fail on complex tasksMost AI agents rely on grep to build context. They don't know what they don't know. They find files but miss architecture. Match strings but lose patterns.THE RESULTAgents that start strong but degrade quickly, requiring constant re-explanation and manual intervention.GETS LOST IN LIMITED CONTEXTYour architectural patternsDependencies across servicesEdge cases buried in legacy codeCoding standards and conventionsRelated files and configurationsThe full picture of what you're building002Why context quality determines code qualityAugment generated pull requests that matched or exceeded human code quality, significantly outperforming competitors.< -500>50Agent UnderperformsHumanAgent OutperformsOverallOverallAggregate performance across functional correctness, style, and context awareness.-13.9-11.8+12.8CorrectnessCorrectnessCode executes as intended, passes tests, and handles edge cases without logical errors.-11.8-9.3+14.8CompletenessCompletenessSolution fully implements the requested feature scope, leaving no placeholders or TODOs.-12.4-12+18.2Code ReuseCode ReuseIntelligently leverages existing project utilities, types, and components to minimize technical debt.-15.8-9.3-4.4Best PracticeBest PracticeCode matches the unique patterns, naming conventions, and architecture of the codebase.-16.4-10.5+12.4Blind study comparing 500 agent-generated pull requests to merged code written by humans on the Elasticsearch repository — 3.6M Java LOC from 2,187 contributorsCursorClaude CodeAugment CodeFULL CODE SEARCHReal-time semantic understandingThe Context Engine is not just grep or keyword matching – it's a full search engine for your code.Augment semantically indexes and maps your code, understanding relationships between hundreds of thousands of files.When you ask "add logging to payment requests," it maps the entire path: React app, Node API, payment service, database, webhook handlers. It then adds logging at every critical point using your existing patterns.> add logging to our payment api> add patch operation to users API> fix checkout timeout issuesunderstood:paymentloggingobservabilityAPI requestscoverage4 of 10 files(40%)matches1 keyword·3 semanticcontext2,847/ 32k tokenssaas-api/srcapi/users.tsapi/payments.ts→ processPayment(amount)94%keywordapi/subscriptions.tsservices/billing.service.ts→ createInvoice(user)89%semanticservices/auth.service.tsmiddleware/validator.tsmodels/user.model.tsutils/stripe-client.ts→ stripe.charges.create()91%semanticlib/telemetry.ts→ logEvent(name, data)87%semanticconfig/database.tsFig. 002—Semantic search retrievalOur Context Engine knows:What's active vs. deprecatedHow services connect and depend on each otherWhat you're actually working on right now in your IDEMore than just your codebaseAugment is grounded in your team's reality. We go beyond syntax, creating agents that write production-ready code and accurate specs because they understand your reality, not just your repository.Commit historyWhy changes were made, not just what changedCodebase patternsHow your team actually builds, not generic best practicesExternal sourcesDocs, tickets, and design decisions via native integrations and MCPTribal knowledgeEdge cases and team conventions discovered through deep codebase analysisINTELLIGENT CONTEXT CURATIONFrom millions of lines to exactly what mattersSignal over noise, automatically.Our Context Engine doesn't dump your entire codebase into the prompt. It:•Retrieves only what matters for your request•Compresses context without losing critical information•Ranks and prioritizes based on relevance•Respects access permissions with proof of possessionResult: The Infinite Context Window. You don't think about tokens. You think about shipping.AugmentOther ToolsActivityWriting codeSearching filesUsing toolsAnalyzing contextSearching filesRe-explaining contextWriting codeSearching filesContext limit reachedRe-explaining contextFig. 003—Code quality over session durationSEE IT IN ACTION60 seconds. One task. Clear winner.augment — context-engine-demoClick to play demoDemo: Context Retrieval on the Elastic Search CodebaseTEAM IMPACTBetter context makes better teamsWhen every developer has instant access to the full context of your codebase, the entire team moves faster and makes better decisions.0%Faster onboardingOne customer reduced engineer onboarding from 18 months to 2 weeks on a legacy Java monolith. New hires ship production code in days.010xFaster refactoringA 150+ person team completed their most complex workflow refactoring in one week. Originally estimated at 6 months, with full test coverage.020hrsSaved monthlyA 200+ person team cut PR review time from 7 minutes to 3 minutes. Senior engineers see 35% higher velocity, spending less time reviewing.030%Fewer production bugsTest coverage increased from 45% to 80% in one quarter without extending timelines. Tests caught edge cases that manual testing missed.04Get StartedGive your AI the context it needs to truly understand your codeCreate your account. The Context Engine works with codebases of any size, from side projects to enterprise monorepos.Install AugmentView PricingWorks everywhereJetBrains, VSCode, and Terminal2-minute setupWorks in your IDESOC 2 compliantEnterprise-ready security --- Skip to contentProductAgentIntentCode ReviewSlackCLIContext Engine MCPChangelogContext EnginePricingDocsBlogResourcesCustomersCareersContextWikiStatus PageTrust CenterSecurityTalk to SalesSign inInstallResourcesBlogRelevant Content for Developers building modern software with AIMar 24, 2026How I went AI-native in my terminal workflowMark ParienteReadMar 16, 2026Review the spec, not the codeSylvain GiulianiReadAll PostsSpecs are infrastructure in the age of agentsMar 13, 2026Specs are infrastructure in the age of agentsMar 13, 2026How we hire AI-native engineers now: our criteria Mar 12, 2026How we hire AI-native engineers now: our criteria Mar 12, 2026How we built a high-quality AI code review agentMar 10, 2026How we built a high-quality AI code review agentMar 10, 2026Building faster than you can thinkMar 9, 2026Building faster than you can thinkMar 9, 2026Context is the product. And now it’s portable.Mar 6, 2026Context is the product. And now it’s portable.Mar 6, 2026Prompts are infrastructure: building agents that actually listenMar 5, 2026Prompts are infrastructure: building agents that actually listenMar 5, 2026GPT-5.4 is now the default model in Augment and free for a limited time. Here’s why.Mar 5, 2026GPT-5.4 is now the default model in Augment and free for a limited time. Here’s why.Mar 5, 2026You don't have an AI problem. You have a transformation problem.Feb 28, 2026You don't have an AI problem. You have a transformation problem.Feb 28, 2026Your agent's context is a junk drawerFeb 27, 2026Your agent's context is a junk drawerFeb 27, 2026The hardest part about going AI-native isn't just technicalFeb 27, 2026The hardest part about going AI-native isn't just technicalFeb 27, 2026Load More (75 remaining)Get StartedGive your codebase the agents it deservesInstall Augment to get started. Works with codebases of any size, from side projects to enterprise monorepos.Install AugmentContact Sales