Dosu for Enterprises
AI-native engineering orgs are built on AI-native knowledge infrastructure. When your agents have the right context, they run faster, cost less, and build correctly the first time. And when that knowledge is accessible to every team (product, QA, support, leadership), the entire organization operates at AI speed, not just engineering.
The knowledge layer your AI agents are missing
Dosu shortens that path dramatically. When you install Dosu, it evaluates how AI-ready your organization actually is — do you have agent-friendly specs, skills documentation, architecture docs, searchable indexes — and surfaces immediate actions it can take to close the gaps. Days, not quarters. You stay in control of what your agents know, how they behave, and where they pull context from. Dosu builds and maintains the knowledge layer. Your team builds the product.
- LangChain
- Hashicorp
- Datadog
- Dify
- Cur8 Capital
What Dosu does for Enterprises
Auto-generate and maintain living documentation.
Dosu connects to GitHub, Jira, Slack, Confluence, and Notion and builds documentation for you. When code changes, docs change. No documentation sprints. No drift.
Make your AI tool investment pay off.
Copilot, Cursor, and other coding agents are only as fast as the context they have. Dosu feeds them current, structured knowledge — making them 2x faster and cutting token usage by 50%.
Cut onboarding time from days to hours.
New engineers typically spend up to 5 days getting enough context to contribute. With Dosu, that drops to under 5 hours — whether you have 50 engineers or 5,000.
Works with the tools your team already uses.
Dosu plugs into your existing engineering stack in minutes. No ripping and replacing the tools your team depends on.
See where AI is working — and where it isn't.
Most enterprise teams adopt AI tools and hope for the best. Dosu gives you the visibility to actually manage it.
Agent performance
See which agents are producing consistent results, which are struggling, and where context gaps are causing failures across every team in your org.
Knowledge bottlenecks
Identify which systems lack documentation coverage and where engineers and agents are asking the same questions repeatedly. Invest in the right context instead of guessing.
Adoption and ROI
Track which teams are getting the most value with AI-native analytics. Measure token savings and time-to-contribution, and report to leadership with concrete data.
Control what your agents know
You decide what knowledge flows where — which teams, which agents, which systems.
How does auth work in this repo?
I need to add a new OAuth provider.
Auth was recently migrated to OAuth2 with PKCE in PR #847. The entry point is src/auth/provider.ts which handles token exchange. To add a new provider:
1// src/auth/providers/newProvider.ts
2export const newProvider = createOAuthProvider
3({ clientId: process.env.NEW_PROVIDER_ID, scopes: ['read',
4'write'], tokenEndpoint: '/oauth/token' })Note: the old JWT flow in legacy-jwt.ts was deleted. All new providers must use PKCE — no implicit flow per team decision in #1180.
- PR #847 — Migrate auth to OAuth
- AGENTS.md — Auth provider pattern
- #1180 — Require PKCE for all providers
- src/auth/provider.ts
- src/auth/middleware.ts
- auth-architecture.md — Token flow docs
Frequently asked questions
Dosu can answer questions or update context from wherever you happen to be (terminal, IDE, browser, GitHub) with full knowledge of your code, specs, docs, issues and open source dependencies.
How long does it take to get Dosu set up?
Does Dosu replace our existing documentation tools?
How do we measure ROI?
Can we control what knowledge agents have access to?
How does Dosu work with Copilot, Cursor, and other AI tools?
Can we start with a specific team before rolling out org-wide?
Ready to fix your documentation problem at scale?
Book a demo and we'll show you how Dosu works inside your specific stack. Or start with a free trial if you'd like to explore on your own first.