AI development crew

From issue to pull request.
On autopilot — with you in charge.

PRKrew turns a GitHub or local issue into a planned, agent-executed pipeline — analyse, review, implement, QA — run by squads of AI agents across every repo in your project, ending in one pull request per repo. No auto-merge, ever.

How it works

An issue goes in. Reviewed pull requests come out.

Every step is visible on a live board — a graph of steps wired like a circuit, each owned by a squad of agents.

01 · ISSUE

An issue lands

From GitHub, Gitea, JIRA — or created locally, fully offline. Labels decide what PRKrew picks up.

githublocal
02 · PLAN

The architect plans

The issue becomes a step-by-step graph, grounded in your actual code. Optionally hold it for your approval before anything runs.

groundedplan gate
03 · SQUADS RUN

Squads execute

Agent squads work steps in parallel across repos — review, implement, QA — with live output, tool traces and token spend on the board.

reviewimplementqa
04 · PR

One PR per repo

Finished work lands as pull requests, ready for human review. You run the result, you complete the merge.

no auto-merge
See it work

Ninety seconds at a time

Short, narrated walkthroughs of the pieces that matter.

0:56

Getting started

A setup checklist connects your AI engine, CLIs and repositories in minutes.

0:52

Human in the loop

Approval gates, a "Needs you" inbox, and runs that hold until you say go.

0:54

Observability & savings

Cost per run, per model, per tool call — and what the work would have cost a human team.

Autonomous, but accountable

You stay in charge

Autonomy you can actually trust in a real codebase.

gate.awaiting

Approval gates

Hold runs at stage boundaries — before QA, before a PR is pushed. Approve from the board, the inbox, or a push notification on your phone.

needs-you

One inbox for everything human

Plans to approve, halted runs, waiting gates, budget alerts — every decision lands in one place and links straight to the work.

$ budget.stop

Budget guardrails

Live USD spend per run with warnings and hard stops. A run can't blow through the ceiling you set.

claude · codex · copilot · gemini

Your CLIs, your models

Runs through the agent CLIs already on your machine, with cross-runtime failover — cloud or fully local models.

workspace → project → repo

Built for multi-repo work

A project spans many repositories; each step targets the repos it needs and fans out one pull request per repo.

run_metric

Measured to the token

Every run is recorded — cost, tokens, duration, failures, tool calls — with per-country rate cards that turn runs into a defensible savings number.

Want to see PRKrew on your codebase?

PRKrew is in active development at KH-IT. Get in touch for a demo or early access.

Get in touch