Skip to content
How I AIHow I AI

Build an AI code review agent with Vercel Eve (full tutorial)

AI writes most of my code now, and that created a new problem: a PR queue I couldn’t keep up with. In this episode, I walk through how I built Merge Mommy, a Vercel Eve agent that reads every PR after checks pass, scores it across six risk dimensions, auto-approves the low-risk ones, and pings me in Slack for anything that needs a human. I built the whole thing in one Codex session, it’s SOC 2 compatible, and it’s already cleared my backlog. *What you’ll learn:* 1. Why AI-generated PRs create a review bottleneck and why the answer isn’t reviewing all of them 2. How Intercom 5x’d PR approval speed and reduced revert rates by putting AI in the review loop 3. Why Vercel Eve is the simplest framework I’ve found for deploying AI agents in Slack and GitHub 4. How I built a full PR review agent in Codex with one prompt and a few steering turns 5. The six components I use to score PR risk (blast radius, reversibility, data security, ops impact, verification gap, and change surface) 6. How I used Chrome browser use to handle Slack bot and GitHub app configuration so I never had to click through setup screens manually 7. Why auto-approved PRs can be SOC 2 compliant as long as the process is auditable, queryable, and in your risk policy 8. How to set up Slack escalation so low-risk PRs become a two-click merge with no manual review *Brought to you by:* WorkOS—Make your app Enterprise Ready today: https://workos.com?utm_source=lennys_howiai&utm_medium=podcast&utm_campaign=q22025 *In this episode, we cover:* (00:00) The PR review backlog problem nobody’s talking about (02:35) Why you don’t have to review every AI-generated PR (05:14) How Intercom built AI-approved PRs (and proved they’re safer) (06:10) How the Eve framework works (directory, skills, channels, connectors) (09:16) The Codex prompt I used to build the entire bot (11:36) What the agent actually does: read, score, approve, or escalate (13:07) Setting up your Eve agent (15:47) The six-component risk scoring model (17:23) Merge Mommy in action: three live PR examples (21:10) Recap and how to build your own version *Tools referenced:* • Vercel Eve: https://vercel.com/eve • Vercel AI SDK: https://sdk.vercel.ai/ • Vercel Chat SDK: https://chat-sdk.dev/ • Codex (OpenAI): https://openai.com/codex *Other references:* • AI is approving our pull requests: Here’s how we made it safe: https://www.intercom.com/blog/ai-is-approving-our-pull-requests-heres-how-we-made-it-safe/ • Review is the bottleneck now: How we let AI approve pull requests (safely): https://rewind.com/blog/ai-approve-pull-requests-safely/ *Where to find Claire Vo:* ChatPRD: https://www.chatprd.ai/ Website: https://clairevo.com/ LinkedIn: https://www.linkedin.com/in/clairevo/ X: https://x.com/clairevo _Production and marketing by https://penname.co/._ _For inquiries about sponsoring the podcast, email jordan@penname.co._

Claire Vohost
Aug 5, 202624mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Auto-approve low-risk PRs using an Eve AI review agent

  1. AI-generated code is creating PR review backlogs, and Claire argues teams can safely stop human-reviewing every PR by using an AI-in-the-loop risk scoring and approval process.
  2. Inspired by Intercom’s results, the approach aims to improve both speed and safety—auto-approved PRs can move faster while maintaining auditability for compliance environments like SOC 2.
  3. The tutorial walks through implementing the agent with Vercel Eve’s directory-based framework (instructions, skills, tools, channels) and Vercel connectors for easy Slack/GitHub integration.
  4. The agent reviews diffs after CI is green, assigns a low/medium/high risk rating via a six-factor model, posts evidence in the PR, and either approves, requests changes, or escalates to humans.
  5. Claire highlights operational realities (repo rules, required human approvals) and recommends running ongoing evals to measure and continuously improve agent decisions.

IDEAS WORTH REMEMBERING

5 ideas

You can reduce PR review load without sacrificing safety by gating approvals on risk.

Instead of reviewing everything, the agent classifies changes as low/medium/high risk, allowing low-risk PRs to be fast-tracked while preserving human attention for risky areas like auth, billing, or API behavior changes.

Compliance concerns are addressable if the process is auditable and policy-backed.

Claire emphasizes that frameworks like SOC 2 can accommodate AI-assisted approvals when risk policies are documented and decisions are traceable (who/what approved, evidence, and logs).

Eve lowers the “agent plumbing” cost by standardizing channels and integrations.

Eve agents are largely a directory of instructions and TypeScript skills, while Vercel connectors handle painful OAuth/refresh-token and channel setup for GitHub and Slack.

Triggering after checks are green is a practical way to cut noise and risk.

The bot waits until CI/actions complete successfully, then reviews the diff and metadata; this avoids premature reviews and makes “verification gap” part of the risk assessment.

A simple, explicit scoring rubric beats vague ‘LGTM’ automation.

The bot scores risk across six factors (blast radius, reversibility, data security, operational impact, verification gap, and category/area risk) and uses thresholds (≤24 low, 25–64 medium, ≥65 high) to drive actions.

WORDS WORTH SAVING

5 quotes

Everybody is drowning in AI-created PRs.

Claire Vo

You don't have to review all your PRs. Yep, you heard it here first. You do not actually have to review one-on-one by a human all those PRs.

Claire Vo

AI-approved PRs and AI-written code in general can be even safer and even higher quality than what you're doing with just a human in the loop.

Claire Vo

I often get this objection from folks. They say, "Well, I'm in a SOC 2 environment. I couldn't possibly auto-approve PRs."

Claire Vo

I like to say this thing where we can put AI to work for us or we can have AI put us to work, and this is an example where you do both.

Claire Vo

PR backlog caused by AI-coded changesIntercom’s AI-approved PRs metrics and compliance framingVercel Eve framework: directory, instructions, skills, toolsGitHub/Slack connectors and multi-channel agent setupCodex prompt + browser-use for SaaS configurationSix-component PR risk scoring model and thresholdsOperational workflow: auto-review + Slack escalation + human merge

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.