Article · 2026-08-05 · 6 moments

Build an AI code review bot in 30 minutes with Vercel Eve

Watch now | 🎙️ I used Vercel Eve agents and Codex to build Merge Mommy: a PR review bot that scores risk, auto-approves the easy ones, and pings me in Slack for the rest ✦ AI generated

Read original ↗

GUEST Claire Vo

timeline · colored by role

Context◆

AI writing most of my code created a PR queue I couldn't keep up with, so the answer isn't reviewing every AI-generated PR but routing them through an agent that scores risk, auto-approves the low-risk ones, and escalates the rest.

Claire introduces the motivating problem and the solution: an AI bot (Merge Mommy) that reviews every PR, scores risk across six dimensions, auto-approves low-risk ones, and only escalates the rest to a human.

transcript

Claire Vo: 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.

explains mechanism · 1supports · 1

Claim

You don't have to review every AI-generated PR because most are low-risk reversible changes that can be safely auto-approved.

Claire argues that the bottleneck of AI-generated PRs is not solved by reviewing everything; most changes are low-risk and reversible, so they can be delegated to the agent.

transcript

Claire Vo: Why AI-generated PRs create a review bottleneck and why the answer isn't reviewing all of them.

extends · 1supports · 2

Data

Putting AI in the review loop can be safer than human-only review, as evidenced by Intercom achieving 5x PR approval speed and reduced revert rates.

Claire cites Intercom's results — 5x faster PR approval and lower revert rates — as evidence that AI-approved PRs can actually be safer than relying solely on human review.

transcript

Claire Vo: How Intercom 5x'd PR approval speed and reduced revert rates by putting AI in the review loop... How Intercom built AI-approved PRs (and proved they're safer).

supports · 1

Mechanism

Vercel Eve is the simplest framework I've found for deploying AI agents in Slack and GitHub because it abstracts configuration through directory, skills, channels, and connectors.

Claire explains how the Eve framework works — via directory, skills, channels, and connectors — and why that makes it the easiest way she has found to ship agents into existing Slack and GitHub workflows.

transcript

Claire Vo: Why Vercel Eve is the simplest framework I've found for deploying AI agents in Slack and GitHub... How the Eve framework works (directory, skills, channels, connectors).

Anecdote

The whole PR review agent was built in Codex with a single prompt and a few steering turns, and using Chrome browser use let me configure the Slack bot and GitHub app without manually clicking through setup screens.

Claire describes the build process: one Codex prompt plus steering turns generated the full agent, and Chrome browser use handled Slack bot and GitHub app configuration so she never had to click through setup manually.

transcript

Claire Vo: How I used Chrome browser use to handle Slack bot and GitHub app configuration so I never had to click through setup screens manually.

gives example · 1

Claim◆

Auto-approved PRs can still be SOC 2 compliant as long as the process is auditable, queryable, and explicitly part of your documented risk policy.

Claire addresses the compliance question directly: automating approvals is compatible with SOC 2 provided the automated process is auditable, queryable, and written into the organization's risk policy.

transcript

Claire Vo: Why auto-approved PRs can be SOC 2 compliant as long as the process is auditable, queryable, and in your risk policy.

provides context · 2

Highlight slides

The Problem: AI Coded Too Much to Review✦ from: AI writing most of my code created a PR queue I couldn't keep up with, so the answer isn't reviewing every AI-generated PR but routing them through an agent that scores risk, auto-approves the low-risk ones, and escalates the rest.The Fix: Merge Mommy, the RisReview Agent✦ from: AI writing most of my code created a PR queue I couldn't keep up with, so the answer isn't reviewing every AI-generated PR but routing them through an agent that scores risk, auto-approves the low-risk ones, and escalates the rest.Auto-Approved PRs Can Still Be SOC 2 Compliant✦ from: Auto-approved PRs can still be SOC 2 compliant as long as the process is auditable, queryable, and explicitly part of your documented risk policy.

Related episodes

Video · My First Million

AI writing most of my code created a PR queue I couldn't keep up with, so the answer isn't reviewing every AI-generated PR but routing them through an agent that scores risk, auto-approves the low-risk ones, and escalates the rest.

You don't have to review every AI-generated PR because most are low-risk reversible changes that can be safely auto-approved.

Putting AI in the review loop can be safer than human-only review, as evidenced by Intercom achieving 5x PR approval speed and reduced revert rates.

Vercel Eve is the simplest framework I've found for deploying AI agents in Slack and GitHub because it abstracts configuration through directory, skills, channels, and connectors.

The whole PR review agent was built in Codex with a single prompt and a few steering turns, and using Chrome browser use let me configure the Slack bot and GitHub app without manually clicking through setup screens.

Auto-approved PRs can still be SOC 2 compliant as long as the process is auditable, queryable, and explicitly part of your documented risk policy.

I spent 24hrs with a blue collar millionaire

simonw/pedalican

Quoting Thibault Sottiaux