DiffGuardianKnow Your CodeUnderstand your code
as fast as your agents write it.
Dozens of agents, hundreds of PRs, and you still have to know what's shipping.
DiffGuardian gives you a spoken walkthrough of your PR and codebase, highlighting the code and building live visuals as it answers. Built for engineers, PMs and anyone building software.
Understand what will be built, FAST.
Your agent wrote a plan. Understand it without reading it.
Agents write plans that run hundreds of lines of markdown, every day. DiffGuardian reads Spec Kit, OpenSpec, Kiro and plain PLAN.md files and turns them into a summary, an architecture diagram, a data model and API tables. Then you ask it questions, by voice or in text, instead of scrolling.
…Grocery Items is the write target: the builder syncs through it, and it has three tasks to build. Nothing is started yet, and the Pantry service is the only new component.
- 01Requirements, extractedWhat the plan promises, what's missing and what it left open, on one screen.
- 02Diagrams, not paragraphsArchitecture on your real folders, plus a data model and API tables.
- 03Progress at a glanceTasks per component: how much there is to build before it starts, and how far along it is after.
Pro. How Plans works
Understand what is being built, FAST.
Watch your agent code. Ask what it is doing. See what it could break.
Point agents at their own git worktrees and DiffGuardian follows each one. Its diff and dependency graph update every few seconds as files change, so you can ask by voice what the agent is up to and see the blast radius while it is still working.
…it just added RefundService, which calls PaymentGateway. That puts the webhook handler in its blast radius, and the new test already covers the service.
- 01Watch it growThe graph of each worktree's changes redraws as the agent writes files.
- 02Ask while it worksHold Space and ask what it is doing, and hear the answer without stopping it.
- 03See the blast radiusWhat its changes reach, before there is a pull request.
Pro. How Worktrees works
Understand what was built by others, FAST.
Get a live walkthrough of any PR, out loud, right now.
Hold Space and ask what a PR does. DiffGuardian walks you through it as it goes: the features it ships, what each one changes, what it touches. You hear it while the code it's explaining scrolls into view and is highlighted. Ask a follow-up whenever you like.
…this hunk replaces the hard-coded admin check with a list of actions for each role, so editors can publish but only admins can delete.
- 01InstantNo setup per PR. Ask and it starts talking.
- 02LiveThe code it's explaining is highlighted and scrolled into view as it talks.
- 03Yours to steerAsk for the part you care about first, or skip to the one file that matters.
Pro. Runs on your own OpenAI or ElevenLabs key, or your system's built-in voice.
Understand any part of your entire repo, FAST.
Ask the whole repo a question. Hear the walkthrough. See it on a live graph.
DiffGuardian, available on macOS, Windows and Linux, indexes your repo into a code knowledge graph. Hold Space and ask how auth works or where a webhook is verified. It answers out loud while a live visual of the flow builds on screen, straight from your call graph.
…then authMiddleware asks PermissionService whether that role can perform the action, and answers with a 403 if it can't.
Every request passes through authMiddleware. It verifies the token, loads the user's role, then asks PermissionService whether that role is allowed to do the action. Admin routes require the admin role. Anyone else gets a 403, and the denial is logged.
▤ Role-based access · on canvas- 01Ask out loudHold Space and ask. No typing, no prompt writing.
- 02Hear it explainedPlain English, with the real files named as it goes.
- 03See the graphThe visual aid builds as it talks.
Prefer typing? The chat is still there.
Prefer to read? Trace It gives you the walkthrough at your pace.
Hit Trace It! and a 5,000-line PR becomes a guided tour, one feature at a time, each explained in plain English. Read it as slowly as you like, go back, and approve having seen the whole thing, not skimmed a diff.
- Walkthrough, written — the PR broken into the features it ships, each explained in plain English
- Blast radius — a dependency graph built from the diff alone (heuristic, not AI): every file and module a change actually touches, before it merges
Want it spoken instead? Talk to your PR.
Ten agents, forty PRs. You still know what's shipping.
When agents work in parallel, the bottleneck is you. For every PR, DiffGuardian finds the quickest way for you to understand it, so you spend minutes on each one instead of reading every line or rubberstamping it.
- Read itA written walkthrough, feature by featureTrace It breaks the PR into what it ships and explains each part at your own pace.See how
- Hear itThe same walkthrough, spokenAsk out loud and get it told back, with the code it's explaining highlighted as it goes.See how
- See itA graph of what a change touchesEvery file and module a change reaches, drawn from the diff alone, before it merges.See how
For everyone who needs to know how their products are actually built.
Working with a team?
Your models. Your keys.
Your voice.
Your code never touches our servers.
Bring your own key for Anthropic, OpenAI, Gemini or OpenRouter, or run a local model. Pick a model for each job. Voice runs on your own OpenAI or ElevenLabs key, or your system's built-in voice, and none of it reaches our servers.
- One OpenRouter key reaches ~400 models, with no new account per vendor
- You pay the provider directly, with no markup
- Ollama, LM Studio, llama.cpp and vLLM work out of the box
- Knowledge bases stay on your machine, and credentials sit in your OS keychain
Your other tools flag things. DiffGuardian helps you understand them, quickly.
| Your PR tools | DiffGuardian, alongside |
|---|---|
| Flag issues and leave comments on the pull request | Explain what the change does, so you can judge the comments |
| Give you text to read | Talk you through it, by voice or in text, highlighting the code as it goes |
| Describe a change in words | Draw what it touches, and build a graph when you ask about the whole codebase |
| Act once the code is written | Also helps before and while it is written: an agent's plan as a summary and diagrams, and its worktree as it codes |
| Use the models the vendor picked, by default | Uses your own models and keys, or fully local, so your code goes only where you send it |
Keep the tools you already use. DiffGuardian sits next to them and gets you to the same understanding in a fraction of the time. It indexes your whole repository into a code knowledge graph and a file index on your machine, so what it tells you is grounded in your real code. See how it fits.
Free to start. Upgrade to talk to your code.
Every plan works with local models. Pro adds voice and cloud providers on your own keys. Enterprise, with single sign-on, is coming soon.
For engineers who want to move fast.
7-day free trial. Cancel anytime.
Try Pro free for 7 days- Everything in Basic
- Voice: talk to your PRs and your codebase
- Unlimited repositories
- Managed AI Providers — BYOK, incl. OpenRouter
- Trace It, Ask, Explain and Polish
- Analyze Project, knowledge bases and RAG
- Plans and Worktrees: understand what agents will build, and are building
- MCP Servers integration
- And many more…
For teams moving fast, working together.
7-day free trial. Cancel anytime.
Register my team- Everything in Pro, for everyone
- Centralised seat billing for the whole team
- Seat management and invoicing
- Minimum of 3 seats
Set up by an organisation admin, who buys the seats for the team. Everyone else just downloads the app — individual sign-in happens inside DiffGuardian during onboarding, never here.
Every feature, side by side.
| Features | BasicFree | ProUSD $8/mo | TeamUSD $8/seat/mo |
|---|---|---|---|
| Connected repositories | 1, public only | Unlimited | Unlimited |
| Models | |||
| Locally-hosted models (Ollama, LM Studio, llama.cpp, vLLM) | |||
| Cloud providers with your own keys (Anthropic, OpenAI, Gemini) | – | ||
| OpenRouter — ~400 models on one key | – | ||
| Review & understanding | |||
| Voice: talk to PRs and your codebase | – | ||
| Manual review comments and Finalize | |||
| Dependency graphs | |||
| Stack maps | |||
| Trace It — guided walkthrough of a diff | – | ||
| Ask — question your project and PRs | – | ||
| Explain & Polish | – | ||
| Analyze Project | – | ||
| Plans — understand an agent's plan | – | ||
| Worktrees — watch agents code and ask about them | – | ||
| Knowledge bases & RAG | – | ||
| MCP Servers | – | ||
| Teams | |||
| Centralised seat billing | – | – | |
| Add or remove users any time | – | – | |