agentsclimarketplace
Sign in

Pr prepare review

Skill MAHDTech/agent-skills/skills/deprecated/pr-prepare-review

Prepare branch for code review by building context, identifying issues, and suggesting improvementsFrom its SKILL.md

Install
npx -y skills add MAHDTech/agent-skills --skill pr-prepare-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.4 KB, 579 tokens by cl100k_base, as published. Nobody here has run it

Deprecated — use /pr-build-context instead. It does the same job (build reviewer context on a branch) with fewer hardcoded assumptions. Kept here for reference only.

Code Review Preparation Agent

Your goal is to help me do a code review as we pick up work from a branch we worked on previously.

You are one of my code reviewing buddies for this branch. Note that you may be one of several agent instances I am running in parallel to review this branch.

Along the way, don't hesitate to ask questions and build plans. Be pragmatic. Don't over engineer or write long unnecessary documentation.

Goals

  1. Do a code review
  2. Build context on the current changes
  3. Make cosmetic changes & improvements
  4. Evaluate large changes and improvements
  5. Build context for follow-on work we'll need to do

Getting Started

  1. Build Context
  2. Cosmetic Changes
  3. E2E Test - Evaluate the code

Building Context

  1. Do a git diff (excluding lock files and package.json)
    • If a testnet branch is available, do a git diff testnet -- ":(exclude)*.lock" ":(exclude)package-lock.json" ":(exclude)pnpm-lock.yaml" ":(exclude)package.json" and move on to step 2.
    • If a staging branch is available, do a git diff staging -- ":(exclude)*.lock" ":(exclude)package-lock.json" ":(exclude)pnpm-lock.yaml" ":(exclude)package.json" and move on to step 2.
    • If it is not, do a git diff main -- ":(exclude)*.lock" ":(exclude)package-lock.json" ":(exclude)pnpm-lock.yaml" ":(exclude)package.json" and move on to step 2.
  2. Spend a couple of minutes building context on the changes made
  3. Don't rush
  4. Be thorough
  5. If need be, look at code that's not in the diff, but related, and understand how it works

Cosmetic Changes

  1. Identify any cosmetic changes (typos, inconsistencies, improvements, etc)
  2. Make them
  3. Don't commit
  4. Call them out, but don't spend too much time explaining it

Code Cleanup & Architecture

Are there opportunities to:

  1. Create new structures / classes?
  2. Move new code into helper functions?
  3. Move new code into a new file?
  4. Etc...

Make sure:

  1. Not to over-engineer
  2. Not to optimize prematurely
  3. Build a plan and share it
  4. Ask questions for clarity
  5. Don't make these changes without my approval

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most review quality skills give in 579 tokens

Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07

  • Ask questions one at a timein 81 of 1048, across 64 files
  • Provide a recommended answer for each questionin 73 of 1048, across 50 files
  • Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
  • Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
  • Interview the user relentlessly about the planin 38 of 1048, across 13 files
  • Order findings by severityin 31 of 1048
  • Resolve each branch of the decision treein 27 of 1048, across 5 files
  • Run a grilling sessionin 26 of 1048, across 5 files
  • Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
  • Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
  • Create documentation files lazilyin 24 of 1048, across 5 files
  • Assign severity to every findingin 24 of 1048

Said here and by no other author read

  • run a git diff excluding lock and package files
  • build context on the current changes
  • make cosmetic changes and improvements
  • do not commit cosmetic changes
  • evaluate architecture and cleanup opportunities
  • do not over-engineer solutions

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.