Agent skill

Create PR Description

by zetaalphavector in zetaalphavector/RAGElo

Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues.

Apache-2.0Auto-check passedDevelopment

Install Create PR Description

skills CLI
$ npx skills add zetaalphavector/RAGElo --skill create-pr-description -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install zetaalphavector/RAGElo create-pr-description --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/zetaalphavector/RAGElo.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/create-pr-description .claude/skills/create-pr-description && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
create-pr-description
GitHub stars
133
Token cost
~1.8k tokens
SKILL.md length
650 words
Files
1
Skills in repo
2
Repo updated
First seen
Licence
Apache-2.0

At a glance

Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues.

  • Works in 6 steps: Gather All Changes → Gather Context → Determine PR Type and Structure → …
  • Create a PR description
  • SKILL.md covers Overview, Output Format, Step 1: Gather All Changes and Step 2: Gather Context, plus 7 more sections
  • Calls git, gh and jq

What it does

Create PR Description is an agent skill from zetaalphavector/RAGElo. Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues. Use when asked to write, generate, or create a PR description, or when preparing a pull request for review.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Pull requests. The repository describes itself as: RAGElo is a set of tools that helps you selecting the best RAG-based LLM agents by using an Elo ranker. The licence is Apache-2.0.

When your agent uses it

  • Create a PR description
  • Preparing a pull request for review

Example prompts

  • “Use the create-pr-description skill to create comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues”
  • “/create-pr-description”

Requirements

  • Python 3

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Gather All Changes
  2. Gather Context
  3. Determine PR Type and Structure
  4. Include Diagrams When Appropriate
  5. Check for Special Conditions
  6. Find Reviewers

What it can do on your machine

Read from SKILL.md and the folder at commit 6f23f0f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git
    • gh
    • jq

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git and gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Create PR Description loads about 1.8k tokens when it runs. Until then it costs about 60 tokens; SKILL.md has 650 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~60
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from zetaalphavector/RAGElo at commit 6f23f0f, republished under its Apache-2.0 licence (© zetaalphavector). 650 words, ~1,843 tokens.

Download SKILL.mdSave it as .claude/skills/create-pr-description/SKILL.md (or your agent's skills folder).
name
create-pr-description
description
Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues. Use when asked to write, generate, or create a PR description, or when preparing a pull request for review.

PR Description Generator Skill

Overview

This skill generates high-quality pull request descriptions for the RAGElo project by analyzing all changes, gathering context, and producing a human-readable summary. RAGElo is a Python library and CLI for evaluating RAG agents using Elo-based tournament ranking, published on PyPI.

Output Format

The agent must produce:

  1. PR Title - A concise, descriptive title (do NOT use conventional commits format like feat(scope): or fix:)
  2. PR Description - Human-readable summary of changes
  3. Proposed Reviewers - 1-2 people based on git history (excluding the PR author)

Step 1: Gather All Changes

Get the list of commits
bash
git log origin/master..HEAD --oneline
Get the full diff
bash
git diff origin/master..HEAD
Get list of changed files
bash
git diff --name-only origin/master..HEAD

IMPORTANT: Do not just read commit messages. You MUST read the actual code changes to understand what was done. Commit messages may be incomplete or misleading.


Step 2: Gather Context

Read CLAUDE.md

Read CLAUDE.md at the repo root for project architecture, commands, and conventions.

This step is required, not optional. Keyword searches often miss relevant issues. Use multiple approaches:

1. Browse recent open issues first (most reliable)
bash
gh issue list --limit 50 --json number,title | jq -r '.[] | "\(.number): \(.title)"'

Scan issue titles manually - the tracking issue for your work may use different terminology than you'd search for.

2. Then search by keywords
bash
gh issue list --search "<keyword1>" --json number,title | jq -r '.[] | "\(.number): \(.title)"'
gh issue list --search "<keyword2>" --json number,title | jq -r '.[] | "\(.number): \(.title)"'

Note: These searches default to open issues. If you need historical context on closed issues, add --state closed explicitly.

3. Check the issue details
bash
gh issue view <issue_number> --json title,body,labels,state | jq '.'

If changes solve or relate to an existing issue, mention it using Closes #123 or Relates to #123 syntax.

Check documentation

Check README.md and docs/ for relevant context and documentation that may need updating.

Check affected code context

For significant changes, read surrounding code to understand:

  • How things worked before
  • Why the change improves the situation
  • Impact on the factory/registry pattern, evaluator hierarchy, or public API

Step 3: Determine PR Type and Structure

For trivial changes (version bumps, typo fixes, simple config changes):

Keep it simple - one or two sentences explaining what changed.

For feature implementations, use What / Why / Impact structure:
markdown
## What
Brief description of what this PR does.

## Why
Explanation of the motivation and context.

## Impact
What parts of the system are affected, any behavioral changes.
For refactoring or architectural changes, use Background / Changes / Migration structure:
markdown
## Background
How things worked before and why that wasn't optimal.

## Changes
High-level description of the new approach.

## Migration
Any manual steps needed (if applicable).
For bug fixes, use Problem / Solution / Testing structure:
markdown
## Problem
Description of the bug and its symptoms.

## Solution
How this PR fixes it.

## Testing
How the fix was verified.
For complex multi-part changes, use Key Changes with bullet points:
markdown
## Summary
One paragraph overview.

## Key Changes
- **Feature A**: Brief description
- **Feature B**: Brief description
- **Refactoring**: Brief description

Step 4: Include Diagrams When Appropriate

For large changes touching multiple components, include mermaid diagrams.

Use diagrams when they genuinely help understanding. Don't add them just for decoration.


Step 5: Check for Special Conditions

Show full SKILL.md (261 more words)Show less
Public API / breaking changes

RAGElo is a published Python library. If the PR changes public-facing APIs (classes, functions, or CLI commands exposed to users), explicitly document:

  1. Whether the change is backward compatible
  2. Any deprecation warnings added
  3. Impact on existing users

⚠️ Breaking Change: This PR modifies the public API. [Describe what changed and migration path]

Security implications

Consider and document:

  • LLM API key handling changes
  • Data handling modifications (user queries, documents)
  • New external integrations or dependencies

Step 6: Find Reviewers

Use git blame to identify people who have recently modified the affected code:

bash
git diff --name-only origin/master..HEAD | while read file; do
  git blame --line-porcelain "$file" 2>/dev/null | grep "^author " | sort | uniq -c | sort -rn
done | grep -v "$(git config user.name)" | sort | uniq -c | sort -rn | head -5

Select 1-2 reviewers who:

  • Have significant contributions to the affected code
  • Are not the PR author
  • Have recent activity

What NOT to Do

❌ Do not create lists of every file that changed with individual explanations ❌ Do not copy-paste commit messages as the description ❌ Do not include implementation details unless critical for understanding ❌ Do not write overly long descriptions for simple changes ❌ Do not skip reading the actual code changes ❌ Do not include the PR author as a reviewer ❌ Do not skip issue checking because "there probably isn't one"

What TO Do

✅ Read all code changes thoroughly before writing ✅ Be concise - readers should understand the change quickly ✅ Focus on the "what" and "why", and to a lesser extent the "how" (unless the "how" is the point) ✅ Include diagrams for complex architectural changes ✅ Mention related issues or documentation ✅ Highlight breaking changes to the public API prominently ✅ Find reviewers who know the affected code ✅ Browse recent issues before writing - the tracking issue may exist under different terminology


Example Output

markdown
# PR Title
Add custom prompt support for retrieval evaluators

# PR Description

Closes #42.

## What
Adds a new `CustomPromptEvaluator` to the retrieval evaluator hierarchy, allowing users to supply their own Jinja2 prompt templates for document relevance evaluation.

## Why
Users with domain-specific evaluation criteria needed a way to customize the LLM prompt without forking the library or subclassing evaluators.

## Key Changes
- **CustomPromptEvaluator**: New evaluator registered via the factory pattern under `RetrievalEvaluatorTypes.CUSTOM_PROMPT`
- **CLI integration**: Added `custom-prompt` subcommand to `retrieval-evaluator` CLI
- **Templates**: Users can pass a Jinja2 template string or file path

## Impact
- No breaking changes to existing evaluators or public API
- New optional dependency on `jinja2` (already a transitive dependency)

---

# Proposed Reviewers
- @alice - Primary author of the evaluator hierarchy
- @bob - Recent contributor to CLI module

© zetaalphavector, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/create-pr-description of zetaalphavector/RAGElo.

Open the folder on GitHubat commit 6f23f0f

Compare with similar skills

Create PR Description next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Create PR Description compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create PR Description this skillzetaalphavector/RAGElo133—~1.8kAutomated safety check: PassApache-2.0
Finishing a Development Branchobra/superpowers297k5 repos~1.9kAutomated safety check: PassMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Check PRonyx-dot-app/onyx32k2 repos~2.3kAutomated safety check: PassMIT
PR Design DocOpenHands/OpenHands90k—~2.4kAutomated safety check: PassMIT
WooCommerce Code Reviewwoocommerce/woocommerce11k3 repos~1.1kAutomated safety check: PassCustom licence

Similar skills

  • Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    297k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed
  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed
  • Check PR

    onyx-dot-app/onyx

    Checks a GitHub, GitLab, or Perforce (p4) pull request (or merge request, or shelved changelist) for unresolved review comments, failing status checks, and incomplete PR descriptions.

    32k GitHub starsUsed in 2 repos~2.3k tokens
    DevelopmentAuto-check passed
  • PR Design Doc

    OpenHands/OpenHands

    For a non-trivial pull request, write a self-contained HTML design doc under the temporary .pr/ directory and link a visibility-appropriate preview in the PR description, so maintainers grasp the…

    90k GitHub stars~2.4k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • WooCommerce Code Review

    woocommerce/woocommerce

    Reviews WooCommerce code changes against the project's standards, flagging backend PHP architecture, naming, documentation, data integrity and testing violations.

    11k GitHub starsUsed in 3 repos~1.1k tokens
    DevelopmentAuto-check passed
  • Record PR Demo

    payloadcms/payload

    A skill your agent uses when a Payload pull request needs a concise visual walkthrough for reviewers.

    45k GitHub stars~1k tokensUpdated yesterday
    DevelopmentAuto-check passed

More from zetaalphavector/RAGElo

  • Review PR

    zetaalphavector/RAGElo

    Reviews pull requests for code quality, pattern conformance, architecture alignment, security, and completeness.

    133 GitHub stars~5k tokensUpdated 16 days ago
    Auto-check passed

Categories

Questions about Create PR Description

What does Create PR Description do?

Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues. Create PR Description is an agent skill from zetaalphavector/RAGElo. Creates comprehensive PR descriptions by analyzing commits, code changes, and gathering context from issues.

When should I use Create PR Description?

Create PR Description fits situations like: create a PR description; preparing a pull request for review.

How do I install Create PR Description in Claude Code?

Run `npx skills add zetaalphavector/RAGElo --skill create-pr-description -a claude-code`. Or copy the skill folder (.agents/skills/create-pr-description in zetaalphavector/RAGElo) into .claude/skills/create-pr-description in your project. Claude Code loads it when a task matches its description.

How do I install Create PR Description in Codex?

Run `npx skills add zetaalphavector/RAGElo --skill create-pr-description -a codex`. Or copy the skill folder (.agents/skills/create-pr-description in zetaalphavector/RAGElo) into .agents/skills/create-pr-description in your project. Codex loads it when a task matches its description.

Can I use Create PR Description in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add zetaalphavector/RAGElo --skill create-pr-description -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-pr-description, .gemini/skills/create-pr-description, .github/skills/create-pr-description and .opencode/skills/create-pr-description in your project.

What does Create PR Description need to run?

Going by SKILL.md and its folder, Create PR Description needs the command-line tools its instructions call (git, gh and jq). Our summary lists: Python 3.

Does Create PR Description access the network?

SKILL.md contains no URLs. Its commands use git and gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Create PR Description safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Create PR Description use?

Create PR Description is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create PR Description use?

About 1.8k tokens (SKILL.md is roughly 7.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Create PR Description?

Skills that share tags, products or a category with Create PR Description: Finishing a Development Branch (obra/superpowers, 297k stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars), Check PR (onyx-dot-app/onyx, 32k stars) and PR Design Doc (OpenHands/OpenHands, 90k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create PR Description?

zetaalphavector (a GitHub organization) maintains it in zetaalphavector/RAGElo, which has 133 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 24, 2026.

Source: zetaalphavector/RAGElo on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.