01src/billing/checkout.ts
Looks like a small price lookup change, but it now writes billing state before Stripe confirms.
Replay checkout success, duplicate webhook, and existing trial renewal before merge.Deterministic PR blast-radius
Every change has a reach.
See it before you merge.
Arbor parses your repo into a call graph and walks every path your PR can actually touch. The answer comes from the graph, not from a model — the same commit always produces the same walk.
First 10 walks freeNo card. See the graph walk on your own pull requests before you pay anything. Then $9 per developer, or $60 flat for a team.
- AST languages
- 9
- code stored
- 0
- agent exports
- 3
- risk card
- share
Example PR walk — GitHub Appcheckout-webhook
This PR looks reasonable, but it can double-charge a retry path that the generated tests never touched.
02src/api/stripe/webhook.ts
Retry worker now runs inside the webhook path and can double-apply invoice side effects.
Add the idempotency regression test the agent skipped.03agent-handoff.md
Codex, Claude Code, or Cursor can receive a scoped repair prompt instead of wandering the repo.
Patch the risky branch first. Stop when the regression test passes.60-second product demo
Watch Arbor turn a tiny diff into a useful review plan.
The point is not another giant report. The point is one PR comment that tells a human and an AI agent where the change reaches, what might break, and which test or patch should happen next.
arbor-botcommented just now
For the solo dev
You get the side effects you did not have time to trace manually: billing, auth, jobs, emails, CLI commands, and shared helpers.
For AI agents
Agents can read the Arbor brief before they keep editing, then limit their next patch to the risky path and generate the missing test first.
For tiny teams
Reviewers see the exact regression to test instead of re-learning the whole codebase in the PR thread.
How it works
From diff to breakage brief in four steps.
Arbor keeps the hard analysis deterministic, then turns the result into review context that humans and agents can both use.
Step 01
Read the PR
Arbor starts from the diff, changed symbols, imports, and framework entry points. It builds the local map reviewers usually reconstruct by hand.
arbor-walk
// PR #184
changed:
- billing/checkout.ts
- billing/plan-id.ts
entrypoints:
- POST /checkout
- POST /webhooks/stripeAgent idea: Arbor as the pre-edit context server.
Before an AI agent edits a repo, it can ask Arbor for the PR's breakage brief. Arbor returns changed paths, likely side effects, unknown edges, and the first test to write. That keeps agent work grounded in the code graph, not just the prompt.
The dashboard
Every answer traces back to an edge you can read.
Pick a pull request and watch the walk redraw. This is the real screen, not a picture of one — and the two plans are two different screens.
1
Hold merge
2
Needs check
2
Ready
idle
Queue
3
Repos watched
Recent walksTrends →
Representative of a real run on a mid-sized TypeScript service. Figures are illustrative, not live data.
Pricing
Half the price of LLM reviewers.
None of the guessing.
Your first 10 walks are free, no card. Deterministic graph walks cost us almost nothing per PR, so they cost you less too — no usage metering, no per-repo surcharge.
Best value
Pro
$9per dev / month
Every PR answered from the graph, not a guess.
Unlimited repos, public and private
Exact breakage paths on every PR — caller by caller
Agent handoff bundles for Claude Code, Cursor, Codex
The first test worth writing, named on every PR
Priority queue — your walks run first
90-day history and shareable risk cards
Reproducible by construction: same commit, same answer
Team
$60per team / month
For codebases where one team's change breaks another's service.
Everything in Pro, for up to 10 developers
Cross-team blast radius — which teams a PR reaches
Ownership routing — reviewers from the graph, not the file tree
Coupling map — which teams are entangled, and trending
Bus factor — load-bearing code with exactly one owner
Coverage gaps — hot files nobody owns
Merge gates — block a PR on its blast radius
1-year history, trends and merge radar
Every install starts with 10 free walks — counted, not timed, and a failed walk never spends one. Pro is per developer; Team is one flat price for up to 10. Annual is $90 and $600. Cancel anytime.
FAQ
Common questions.
How Arbor walks PRs, how agents can use the brief, and how it fits a small-team workflow.
Can agents use it anyway?
Yes. Arbor produces the grounded brief; an agent can use that brief to write tests or keep a patch scoped.
Does Arbor run my code?
No. The worker parses and walks source structure. It does not execute your app, tests, install scripts, or runtime commands.
What if the map is incomplete?
The missing edge is named in the PR so a reviewer or agent knows where the uncertainty is.