npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@howaboua/pi-subagent-review

v0.2.19

Published

Pi extension that adds /review via an isolated review subagent.

Downloads

2,201

Readme

@howaboua/pi-subagent-review

Adds /review, which sends the current repository state to an isolated review subagent and injects its findings back into the main session for triage.

Install

pi install npm:@howaboua/pi-subagent-review

Run /reload after installation if Pi is already open.

Usage

/review
/review focus on migrations and tests
/review loop
/review stack=main@origin
/review stack="description(exact:'release base')" focus on migrations

Anything after /review becomes additional reviewer guidance. A leading loop starts review-loop mode and is removed from that guidance. In a JJ workspace, stack=<ancestor revset> selects a cumulative review base and is removed from that guidance; quote a revset containing spaces. Only one stack= argument is accepted.

Findings are advisory. The command tells the main agent to verify and categorize them against the current implementation and session context rather than treating them as a TODO list.

Review scope

In a JJ workspace, the extension reviews the active revision before considering an enclosing Git checkout. It pins the active change and commit IDs, reviews against its direct parent by default, and uses stack=<ancestor revset> for a cumulative range. The active revision must be conflict-free. The read-only command reviews stored commit content; capture filesystem edits with JJ before invoking /review.

Elsewhere, the extension chooses among local dev, main, and master branches, then reviews from the merge base so committed and dirty changes are included without base-only commits. With no usable base it reviews the checkout; with no changes it reviews the latest commit.

When enabled, a separate model summarizes the current Pi branch for the reviewer; raw turns are not sent. Review continues diff-only if summarization fails.

Review loops

/review loop records a review-specific marker. The next /review summarizes fixes since that point with the configured summary model, advances the marker, and starts another pass. It does not conflict with pi-auto-trees' /marker.

Configuration

On first load, the extension creates ~/.pi/agent/pi-subagent-review.json, or the equivalent path under $PI_CODING_AGENT_DIR.

{
  "model": "openai-codex/gpt-5.6-sol",
  "thinking": "medium",
  "summary": {
    "enabled": true,
    "model": "openai-codex/gpt-5.6-luna",
    "thinking": "low"
  }
}

Models use Pi's provider/model format. Thinking levels through max are accepted and clamped. If a configured model is unavailable, the command falls back to the current session model.

The summary model prepares both reviewer conversation context and review-loop increment summaries.

Do not load another extension that registers /review unless the command collision is intentional.