pr-draft-summary
Generates a draft pull request title, description, and branch name suggestion after implementation work completes.
Identity proven by GitHub's verified organisation badge (openai)
- Agents
- Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, claude.ai and API
- License
- MIT (permissive)
- Pin
- 05241c7pinned Oct 9, 2026
- Context cost
- about 1,800 tokenswhen the skill loads
- GitHub stars
- 3,899on openai/openai-agents-js
- Scripts
- None found
What it does
“Prepare a local PR title, description, and branch suggestion when requested or after eligible implementation work.”
Generated summary from the skill's own text
Generates a draft pull request title, description, and branch name suggestion after implementation work completes. The skill inspects the actual code diff, determines change type from file paths, and produces GitHub-ready copy following repository conventions for the openai-agents-js project.
When to use it
- After eligible implementation work completes and verification is done
- When code, tests, examples, or behavior-impacting docs change
- To prepare PR metadata before creating a branch or pushing changes
- When user explicitly requests a PR draft at any stage
When not to
- No code changes detected in the working tree or commits
- Only repository metadata or documentation without behavior impact changed
- User explicitly opts out of draft generation
Install
These commands install from the publisher's repository and need no account.
skills CLI
npx skills add openai/openai-agents-js --skill pr-draft-summaryInstalls from the repository's default branch, which may have changed since the commit we checked.
GitHub CLI 2.90 or later
gh skill install openai/openai-agents-js pr-draft-summary@05241c7aa3d5629acc7407dd8ee1105637fe0369Installs the commit we checked (05241c7).
Copy by hand
Download the folder at 05241c7 and put it in .claude/skills/pr-draft-summary for Claude Code and in .agents/skills/pr-draft-summary for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.
Copy for your agent
A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.
Prompt
Install the agent skill "pr-draft-summary" by OpenAI. Source, pinned to commit 05241c7aa3d5629acc7407dd8ee1105637fe0369: https://github.com/openai/openai-agents-js/tree/05241c7aa3d5629acc7407dd8ee1105637fe0369/.agents/skills/pr-draft-summary 1. Download that folder at exactly this commit. Do not use the default branch. 2. Copy every file unchanged, keeping the folder structure, into .claude/skills/pr-draft-summary/ for Claude Code and .agents/skills/pr-draft-summary/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist. 3. Do not edit the files, and do not run any script in the folder unless I ask you to. 4. When you are done, list the files you copied and where you put them. Listing and checks: https://www.agentnative.dev/skills/openai/pr-draft-summary
SKILL.md at the pin
---
name: pr-draft-summary
description: Prepare a local PR title, description, and branch suggestion when requested or after eligible implementation work.
---
# PR Draft Summary
## Purpose
Produce the PR-ready summary required in this repository after eligible work is complete: a concise change summary plus a PR-ready title and draft description for openai-agents-js.
## Close-out Gate
1. Inspect the actual task diff before sending the final response. An explicit request for a PR draft takes precedence over the automatic-trigger exclusions below. Producing a draft never authorizes creating a branch, committing, pushing, or opening a PR.
2. Run this skill when the diff includes runtime code, tests, examples, build/test configuration, or docs with behavior impact. Do not use perceived change size to skip an eligible change.
3. Run it after any required verification and changeset work and before sending the "work complete" response.
4. Skip only when no eligible files changed, every change is repo-meta or docs-only without behavior impact, the task is conversation-only, or the user explicitly opts out.
## Inputs to Collect Automatically (do not ask the user)
- Current branch: `git rev-parse --abbrev-ref HEAD`.
- Working tree: `git status -sb`.
- Untracked files: `git ls-files --others --exclude-standard` (use with `git status -sb`; `--stat` omits them).
- Changed files: `git diff --name-only` (unstaged) and `git diff --name-only --cached` (staged); sizes via `git diff --stat` and `git diff --stat --cached`.
- Pull request base reference (use `origin/main`; never use a feature branch's upstream as the PR base):
- `BASE_REF=origin/main`; if it does not exist locally, use `main`.
- `BASE_COMMIT=$(git merge-base "$BASE_REF" HEAD)`.
- Committed branch diff: `git diff --name-only "${BASE_COMMIT}..HEAD"` and `git diff --stat "${BASE_COMMIT}..HEAD"`.
- Commits ahead of the base fork point: `git log --oneline --no-merges ${BASE_COMMIT}..HEAD`.
- Category signals for this repo: runtime (`packages/`, `examples/`, `helpers/`, `scripts/`), tests (`packages/**/test`, `integration-tests/`), docs (`docs/`, `README.md`, `AGENTS.md`, `.github/`), build/test config (`package.json`, `pnpm-lock.yaml`, `pnpm-workspace.yaml`, `tsconfig*.json`, `tsc-multi.json`, `eslint.config.*`, `vitest*.ts`).
## Task ownership
Use the complete task-owned diff, including committed, staged, unstaged, and untracked changes. Inspect the branch diff as context, then identify ownership from the user's request and the task's initial state and implementation history. Do not include unrelated pre-existing branch or workspace changes in the draft. A request covering the entire branch includes that branch's complete diff. If ownership is unclear and would change the draft materially, resolve that concrete question before presenting it as ready.
For a topic-scoped correction on an unrelated feature branch, describe the correction and suggest an appropriate branch name as text only. Preserve the actual checkout and branch. Do not inherit an unrelated issue-closing reference from the current branch name.
## Workflow
1. Run the commands above without asking the user; compute `BASE_REF`/`BASE_COMMIT` first so later commands reuse them. Compare against `origin/main` or `main`, not the current branch's upstream.
2. Combine the task-owned committed diff with task-owned staged, unstaged, and untracked changes. If that combined diff is empty, reply briefly that no code changes were detected and skip emitting the PR block.
3. Infer change type from the touched paths listed under "Category signals"; classify as feature, fix, refactor, or docs-with-impact, and flag backward-compatibility risk only when the diff changes released public APIs, external config, persisted data, or wire protocols. Judge that risk against the latest release tag, not unreleased branch-only churn.
4. Summarize changes in 1–3 short sentences using the most relevant task-owned paths and stats from the committed, staged, and unstaged diffs. Explicitly call out untracked files because `--stat` does not include them. Use commit messages as supporting context, not as a substitute for inspecting the committed diff.
5. Choose the lead verb for the description: feature → `adds`, bug fix → `fixes`, refactor/perf → `improves` or `updates`, docs-only → `updates`.
6. Suggest a branch name. If already on a branch for this task, keep it; otherwise propose `feat/<slug>`, `fix/<slug>`, or `docs/<slug>` based on the primary area (for example `docs/pr-draft-summary-guidance`).
7. If the current branch belongs to this task and matches `issue-<number>` (digits only), keep that branch suggestion. When an issue number is present, use the native same-repository reference `#<number>` and include an auto-closing line such as `This pull request resolves #<number>.`. Do not add the explicit issue URL or wrap the reference in a Markdown link. Do not block if the issue cannot be fetched.
8. Draft the PR title and description using the template below. Apply the repository-wide GitHub paste-readiness rule: use exactly `#123` for same-repository issues or PRs and `owner/repo#123` for cross-repository references; never emit `[PR #123](https://github.com/owner/repo/pull/123)`, `[#123](...)`, Codex navigation links, local file links, Codex-only citation markers or footnotes, or app directives in the copy-ready block. Preserve ordinary descriptive links to API docs, design notes, and other targets without native GitHub issue or pull-request syntax.
9. Normalize references before returning the block: replace every same-repository URL or `openai/openai-agents-js#<number>` reference with `#<number>`, replace every cross-repository issue or pull-request URL with `owner/repo#<number>`, then rescan the full block. Do not return it while a Markdown-linked issue or pull-request label, a same-repository qualified reference, or a bare GitHub issue or pull-request URL remains.
10. Output only the block in "Output Format". Keep any surrounding status note minimal and in English.
## Output Format
When closing out a task, add this concise Markdown block (English only) after any brief status note unless the task falls under the documented skip cases or the user says they do not want it.
```
# Pull Request Draft
## Branch name suggestion
git checkout -b <kebab-case suggestion, e.g., feat/pr-draft-summary-skill>
## Title
<single-line imperative title, which can be a commit message; if a common prefix like chore: or feat: etc., having them is preferred>
## Description
<include what you changed plus a draft pull request title and description for your local changes; start the description with prose such as "This pull request resolves/updates/adds ..." using a verb that matches the change (you can use bullets later), explain the change background (for bugs, clearly describe the bug, symptoms, or repro; for features, what is needed and why), any behavior changes or considerations to be aware of, and you do not need to mention any tests you ran.>
```
Keep it tight—no redundant prose around the block, and avoid repeating details between `Changes` and the description. Tests do not need to be listed unless specifically requested.
From openai/openai-agents-js at 05241c7, shown as plain text under its license: MIT (permissive). The license text is in LICENSE.
View SKILL.md on GitHub at 05241c7Files
- SKILL.md7.3 KB · b4386e31c017
- agents/
- openai.yaml251 B · 31dc54e1d3bc
2 files, 7.5 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.
Skill folder on GitHub at 05241c7More from OpenAI
code-change-verification
Runs formatting, linting, type checking, and test verification for SDK code changes after review.
Contains scripts
Codex · about 1.3K tokens · pinned 7 days ago
docs-sync
This skill audits English SDK documentation by comparing features and configuration options in the codebase against the current docs, then proposes or…
Claude Code, Codex, Cursor +4 · about 1.2K tokens · pinned 7 days ago
examples-run-analysis
Analyzes logs and artifacts from manual examples runs to validate example behaviors, identify failures, and classify issues.
Claude Code, Codex, Cursor +4 · about 1K tokens · pinned 7 days ago
final-release-review
Audits Python SDK release candidates or plans against the previous release, comparing diffs for regressions, API changes, and version compatibility.
Contains scripts
Claude Code, Codex, Cursor +3 · about 5.4K tokens · pinned 7 days ago
implementation-final-review
This skill guides agents through a structured review process for completed implementation changes before final verification.
Contains scripts
Claude Code, Codex, Cursor +3 · about 1.9K tokens · pinned 7 days ago
implementation-kickoff
This skill manages an isolated implementation workflow using a dedicated git worktree, keeping the user's original checkout unchanged while building a…
Contains scripts
Codex · about 2.9K tokens · pinned 7 days ago
Related in Coding workflow
planning
This skill orchestrates multi-workflow projects with dependencies and shared data tables by creating a task graph in the orchestrator via the create-tasks tool.
by n8nOfficial publisher
Claude Code, Codex, Cursor +3 · about 2.4K tokens · pinned 7 days ago
progressive-building
This skill implements incremental workflow development for agents building or extending workflows.
by n8nOfficial publisher
Claude Code, Codex, Cursor +3 · about 1.4K tokens · pinned 7 days ago
workflow-builder
Builds and edits n8n workflows using the TypeScript SDK.
by n8nOfficial publisher
Claude Code, Codex, Cursor +3 · about 14.2K tokens · pinned 7 days ago
remotion-upgrade
This skill upgrades Remotion and its related packages to the latest stable version.
by RemotionOfficial publisher
Claude Code, Codex, Cursor +3 · about 460 tokens · pinned 7 days ago
remotion-upgrade
This skill upgrades Remotion and related packages across a project, automatically detecting the package manager and workspaces.
by RemotionOfficial publisher
Claude Code, Codex, Cursor +3 · about 460 tokens · pinned 7 days ago
implementation-strategy
This skill guides agents through a structured decision-making process for SDK behavior changes and API modifications.
by OpenAIOfficial publisher
Claude Code, Codex, Cursor +4 · about 3.6K tokens · pinned 7 days ago
Open roles at OpenAI
5 of 239 open roles on the jobs board
- Software Engineer, Productivity - Networking
San Francisco, California, United States
- Networking Operating System Firmware Engineer
San Francisco, CA, United States (Hybrid)
- Software Engineer, Private Computing
San Francisco, California, United States (Hybrid)
- Data Center Hardware Quality & Reliability Engineer
San Francisco, California, United States (Remote)
- Software Engineer, Habitat (Online Data)
Seattle