Question 1

Why not just give Claude Code or Cursor my repo directly?

Accepted Answer

Honestly: on a small, well-organized repo, a bare agent does fine. doc0 earns its keep when the repo outgrows a context window, and the moment anyone besides that one engineer, human or agent, needs the same answers. Agents reading raw files rebuild your codebase's structure from scratch every session, burning tokens and inventing function names when context runs out. doc0 builds the structure once (symbols, call graphs, execution flows) and serves it to every tool over MCP. The same index also answers your PM in Linear and your support team in Slack.

Question 2

How is this different from DeepWiki?

Accepted Answer

DeepWiki gives you a wiki to read. doc0 gives you one to keep: it updates as you push, your edits survive regeneration, and one repo gives you both a technical wiki and a user guide. You also get private repos, team roles, custom domains, and an MCP server your AI tools query directly. In our published benchmark, blind judges scored doc0 ahead of DeepWiki on every one of 10 repos tested. Methodology and raw scores are public.

Question 3

How is this different from Mintlify or GitBook?

Accepted Answer

Mintlify and GitBook are editors: you write the docs, they publish them. doc0 writes the docs from your code, keeps them current as you push, and lets you edit on top. It parses your code into a symbol graph (functions, call chains, API endpoints), so the same source also answers your AI tools over MCP. No MDX files to write, no CI pipeline to maintain.

Question 4

Why not just run it locally?

Accepted Answer

Local code-graph tools re-index on every machine and only help the one engineer running them. doc0 is hosted (no per-machine indexing), always fresh (it auto-updates as you push), and shared (one graph serves Slack, Linear, and every AI tool your team uses). On top of the graph you also get a readable wiki, something a local indexer doesn't produce.

Question 5

Does it work with private repositories?

Accepted Answer

Yes. doc0 connects through a GitHub App. Your repository is cloned, analyzed in memory, and deleted after processing. We don't keep a copy. To write and update your docs we store indexed chunks of your code (encrypted at rest, deleted with your project) plus file hashes for change detection, and your code is never used for training. Private repositories are available on Pro, Team, and Enterprise.

Question 6

What happens to my edits when docs regenerate?

Accepted Answer

Your edits stay. doc0 tracks which source files each page came from, so a push regenerates only the pages your change affects. When a change touches text you edited, doc0 keeps your text and flags the page so you can review the update.

Question 7

How do you keep my code safe?

Accepted Answer

Your code is used for inference only and never to train models, and the AI models that generate your documentation keep no copy of it, on every plan. Sensitive account and project changes are recorded in an audit log, and SOC 2 and ISO 27001 certification is underway. The privacy policy lists every sub-processor and where it processes data.

Question 8

How does doc0 work with Claude Code, Cursor, Slack, and Linear?

Accepted Answer

Every doc0 project has an MCP server. Add it to Claude Code, Cursor, VS Code, Codex or Windsurf from the project's Connect agents page: public docs need no account, and private docs ask you to sign in the first time your client connects. Beyond search, agents get structural tools raw file access can't offer: impact analysis, path tracing between symbols, detection of open PRs likely to collide, and who-knows expertise mapping. The Slack bot and the Linear agent answer from the same index on the Team plan: mention doc0 on any Linear issue, or ask it in any Slack channel.

Question 9

How do I make my agent actually use doc0?

Accepted Answer

Installing an MCP server doesn't mean an agent uses it: left alone, agents drift back to searching raw files. Run npx @alessandro-afloarei/doc0 setup in your repo: it detects Claude Code, Cursor, Codex, OpenCode, Gemini CLI or Antigravity and installs three things per agent: the MCP server entry, an always-on rule that points the agent at your docs first, and a skill with triggers tuned to your project. npx @alessandro-afloarei/doc0 remove undoes exactly that. Public docs need no account; for private docs, copy the command from your project's Connect agents page, and each agent asks you to sign in the first time it connects.

Question 10

Does doc0 stop my agent from writing duplicate code?

Accepted Answer

It makes duplicates findable before they're written. Agents query the knowledge graph over MCP and get the existing implementation with its callers, so "write a new helper" becomes "use the one that exists." doc0 doesn't gate your agent's writes (no tool truthfully claims to), but an agent that sees the codebase's structure stops recreating it.

Question 11

What languages and frameworks are supported?

Accepted Answer

doc0 works with any GitHub repository regardless of language. It parses 14 grammars with tree-sitter (JavaScript, TypeScript, TSX, Python, Java, C, C++, C#, Go, Rust, Kotlin, PHP, Ruby, and Swift) and falls back to file-structure analysis for the rest. It detects endpoints in Express, NestJS, FastAPI, Django, Spring Boot, Rails, and 10+ more frameworks.

Question 12

Where does my code get processed?

Accepted Answer

Your repository is cloned, analyzed in memory, and deleted after processing. We keep file hashes, the indexed chunks we use to write and update your docs, and their embeddings, never a copy of your repository. Your code is never used for training. Generation, search and chat use third-party AI sub-processors; our full sub-processor list, with where each one processes data, is in the privacy policy.

Question 13

How much does it cost?

Accepted Answer

Pro and Team start with a 3-day trial. Pro (€30/mo) covers 500,000 lines of code a month, unlimited public projects, 5 private repos, custom domains, and export. Team (€250/mo) adds unlimited team members, 20 private projects, deep research, plus the Slack bot & Linear agent. Enterprise is priced per organization. Book a call.

Question 14

How do you measure documentation quality?

Accepted Answer

We publish our benchmark: blind LLM judges score doc0-generated wikis against DeepWiki and Google CodeWiki on 10 real repositories. Methodology, corpus, and raw scores are public in the benchmarks repository.

Docs written from your code that stay current as you push.

Your codebase shouldn’t live in one person’s head.

Onboarding

Changing code

Trusting the docs

From your repository to docs that stay current.

Connect your repo

doc0 maps your code

Your wiki is written

It stays current

See it on a real codebase.

dub

next.js

mem0

hono

drizzle-orm

fumadocs

Understand your app without reading the code.

Written from your code

Current as you push

Cited to the line

Yours to edit

Diagrams of how it fits

Your docs come from a map of your code, not a guess.

Walkthroughs come from the map

Names checked against your code

Quotes copied from your lines

Your team asks in Slack and Linear. Your agents ask from the editor.

Answers in Slack and Linear

Agents set up with one command

Answers show their sources

See what a pull request changes before you read the diff.

Measured on public code you can check.

Start with a 3-day trial

Pro

Team

Enterprise

Connect your repository, and check the wiki it writes.

No copy of your repository

Never used to train models

You choose who sees what

Questions before you connect a repo