Pr Description logo

Pr Description

OrganizationPopular
TanStack
pr-description

Use when writing a pull request title or body, when about to run gh pr create, when about to git push on a branch that already has an open PR, or when the user says /pr-description, "write the PR description", or "update the PR title". Don't use for commit messages, changelogs, or review comments.

Overview

PublisherTanStack
Repositoryai
Skill namepr-description
Stars
3.2K
Forks
361
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by TanStack on GitHub. Read the source before you install it.

Installation

Install the Pr Description AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/TanStack/ai/tree/main/.agents/skills/pr-description
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/TanStack/ai.git /tmp/ai
mkdir -p .claude/skills
cp -r /tmp/ai/.agents/skills/pr-description .claude/skills/pr-description
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pr Description in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Pr Description on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Pr Description is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

pr-description

Write the GitHub PR title and body, then post them. Do not wait for approval.

Work from the diff, not from memory.

When to run

This skill is auto plus on demand.

Run it:

  • When the user asks for a PR title or body, or types /pr-description.
  • Immediately before gh pr create.
  • Immediately after an agent git push on a branch that already has an open PR. Then run gh pr edit with a fresh title and body.

Do not run it on a human-only git push in another terminal. Do not add a git hook.

A later push must rewrite the GitHub text. Do not leave a stale description.

A fix PR must pass bugfix-pr first. If those gates fail, stop. Do not run gh pr create. Do not run gh pr edit.

Required skills for the title and body

simple-english and i-have-adhd are prerequisites. They apply to the PR title and body, not to the rest of the session.

Load both immediately before writing the title and body. Use the Skill tool if this harness has one. If it does not, Read:

  • .claude/skills/simple-english/SKILL.md
  • .claude/skills/i-have-adhd/SKILL.md

Copies also live at .agents/skills/ (Codex) and .grok/skills/ (Grok). Keep those three files identical.

If either file is missing, stop. Do not post a weaker substitute.

When i-have-adhd is loaded from this skill, ignore its persistence section. Do not switch the rest of the session into ADHD mode.

Use pragmatic simple-english. Then shape the text with i-have-adhd: next action first in the lead, numbered steps for manual test, lists capped at 5.

Procedure

1. Ground in the repo

  1. Find the base branch (main / master / develop).
  2. Read git log --oneline base...HEAD and git diff base...HEAD.
  3. Read .github/pull_request_template.md if it exists (also check .github/PULL_REQUEST_TEMPLATE.md and .github/PULL_REQUEST_TEMPLATE/).
  4. Read about 15 recent PR titles in this repo (gh pr list --limit 15 --state all --json title). Match that shape. Do not invent a new title style.
  5. Collect linked issues from branch name, commit messages, and Closes # / Fixes # text.

2. Classify the PR

Pick one kind from the diff:

KindSignal
fixRestores broken behavior. A user-visible bug, a failing test, a regression.
featAdds behavior that did not exist. New export, new command, new user-facing flow.
chore / docsAgent files, CI, docs-only, refactors with no user-visible behavior change.

If both a feat and a fix are in the diff, the larger user-visible story wins. Say so in the lead.

3. Feat gate: easy test path

  • A new or updated automated test that covers the new behavior
  • A command a reviewer can copy and run
  • A change in an official example app that shows the new behavior

If none of those exist, stop. Do not run gh pr create. Do not run gh pr edit. Tell the user what is missing. Wait until they add one, then continue.

A fix does not use this gate. A fix uses bugfix-pr. Load it first. A chore / docs PR does not use this gate.

4. Write the title

Match this repo's recent PR titles. In a conventional-commit repo, write type(scope): summary. Keep it one line.

The title must still make sense if the reviewer never opens the body.

5. Write the body

Load simple-english and i-have-adhd, then write in this order.

Lead (required, 1 to 4 sentences, before the template).

  • feat: what the PR does for a user of the product.
  • fix: what the bug is, then how this PR fixes it.
  • chore / docs: what changed, in product or repo terms, not a file list.

No preamble. No "This PR aims to". Start with the fact.

Repo template (required when the file exists).

Fill every section honestly. Leave a checkbox unchecked when the claim is false. Do not tick "I ran pnpm test:pr" if you did not run it.

Extra sections after the template, in this order. Testing and Risk / rollback are always present. For a fix, Root cause and Possible alternatives are also always present. Linked issues and Public API change are omitted when they do not apply.

Root cause (fix only)

Required for a fix. Copy from the bugfix-pr report. Do not paste the diff. Write:

markdown
## Root cause

**Issue.** What is broken, for whom, under which inputs.
**Cause.** Why it happens in the code. Name the function or path.
**Fix.** How this PR's change kills that cause.
Possible alternatives (fix only)

Required for a fix. Copy from the bugfix-pr report. Write:

markdown
## Possible alternatives

- **<name>.** What it is. Why this PR did not take it.

If there is no other real way, write None. Do not omit the heading.

Testing

Three parts, all required:

  1. Commands run. What you ran, what passed, what you skipped and why. If you ran nothing, say so.
  2. Manual test. Numbered steps a reviewer can follow to see the change. For a fix, step 1 is how to reproduce the old bug, then how to confirm it is gone.
  3. How this PR makes testing easy. Name the artifacts on the branch: tests, a command, an example app change. If there are none (allowed for fix and chore / docs), write that. Do not invent an easy path.
Linked issues

Omit this heading when nothing links. When an issue links, use Closes #N or Fixes #N.

Risk / rollback

What could break. How to undo (revert the PR, turn a flag off, and so on). If risk is low, say that in one line. Do not skip the heading.

Public API change

Omit this heading when the PR does not touch the published surface.

Published surface means: exported functions, types, components, CLI flags, env vars, and documented contracts from published packages. Tests, internal files, agent skills, AGENTS.md, and CLAUDE.md do not count.

When it does touch that surface, show caller usage, not the internal diff:

markdown
## Public API change

**Before**

\`\`\`ts
old call site
\`\`\`

**After**

\`\`\`ts
new call site
\`\`\`

Two short snippets. What a user writes today, then what they write after this PR.

6. Post immediately

Do not wait for the user to approve the text.

  • No PR yet: gh pr create --title "..." --body-file ...
  • PR already open: gh pr edit --title "..." --body-file ...

If gh fails, stop and show the error. Leave the drafted body in the chat so it is not lost.

Do not attach screenshots to the PR. Do not commit screenshot files for this skill.

Red flags

You catch yourselfDo instead
Writing the body from the branch nameRead the diff.
Ticking a template checkbox you did not doLeave it unchecked and say so in Testing.
Posting a feat with no test, command, or exampleStop. The feat gate failed.
A 1-4 sentence lead that lists filesRewrite as what a user can do, or what bug is gone.
Public API section that shows the internal diffShow caller usage before and after.
Public API section on an AGENTS.md-only PROmit the heading.
"I'll update the description later" after a pushRewrite now with gh pr edit.
Waiting for the user to approve the textPost. This skill does not wait.
Uploading or committing screenshots for the PR bodySkip. Do not put images on the PR.
Writing without loading simple-english and i-have-adhdLoad both from the repo copies. If missing, stop.
Posting a fix before bugfix-pr gates passStop. Load bugfix-pr. Do not create or edit.
A fix PR with no Root cause or Possible alternativesCopy both sections from the bugfix-pr report.

Frequently asked questions

What does the Pr Description AI skill do?

Use when writing a pull request title or body, when about to run gh pr create, when about to git push on a branch that already has an open PR, or when the user says /pr-description, "write the PR description", or "update the PR title". Don't use for commit messages, changelogs, or review comments.

Why use Pr Description on TypingMind?

Because you install it once and use it with any model. Pr Description is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Pr Description in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/TanStack/ai/tree/main/.agents/skills/pr-description. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Pr Description?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Pr Description?

As many as you like. As long as a model supports skills, you can use Pr Description with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Pr Description AI skill free?

Yes. It is published on GitHub by TanStack under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇