pr-draft-summary
Create the required PR-ready summary block, branch suggestion, title, and draft description for openai-guardrails-python after runtime, tests, examples, build/test configuration, or behavior-impacting docs change.
- License 许可
- MIT
- 最近更新
- 2026-11-11
- Stars
- ★ 267
- Forks
- 53
01 前言 Front Matter
- name
- pr-draft-summary
- description
- Create the required PR-ready summary block, branch suggestion, title, and draft description for openai-guardrails-python after runtime, tests, examples, build/test configuration, or behavior-impacting docs change.
02 资源文件 1
- 参考文档agents/openai.yaml251 B
03 SKILL.md 正文
PR Draft Summary
Purpose
Produce a concise, copy-ready branch suggestion, PR title, and description for OpenAI Guardrails Python. This skill creates text only. It never authorizes creating a branch, committing, pushing, opening a pull request, or mutating GitHub.
When to trigger
- Run after applicable final review and verification when the task changed
src/guardrails/,mcp_server/,tests/,examples/, build/test configuration, or behavior-impacting docs. - Run for eligible local-only or uncommitted work even when the user did not ask to open a pull request.
- Skip for repository metadata, editorial docs, conversation-only work, or when the user explicitly asks not to include a PR draft.
Inputs to collect automatically
- Current branch:
git branch --show-current. - Repository state:
git status --short. - Untracked files:
git ls-files --others --exclude-standard. - Staged and unstaged paths and statistics.
- Base reference: the branch upstream when configured, otherwise local
main, otherwiseorigin/main. - Merge base and commits ahead of that base.
- Latest local release tag when compatibility context matters; label it potentially stale when remote tags were not refreshed.
- Category signals:
- Runtime:
src/guardrails/,mcp_server/. - Tests:
tests/. - Examples:
examples/. - Docs:
docs/,mkdocs.yml. - Build/test configuration:
pyproject.toml,uv.lock,Makefile,.github/.
- Runtime:
Do not ask the user for information that can be derived from the repository.
Workflow
- Resolve the base and inspect committed, staged, unstaged, and untracked task content.
- If there are no task changes and no commits ahead of the base, report that no code changes were detected and do not emit the PR block.
- Classify the work as feature, fix, refactor/performance, docs-with-impact, or repository tooling.
- Flag backward-compatibility risk only when the diff changes a released public API, external configuration, persisted data, serialized state, or wire protocol.
- Summarize the complete change in one to three sentences. Include untracked deliverables because ordinary diff statistics omit them.
- Choose a branch name:
- Keep the current non-
main, non-detached branch when it already describes the task. - Otherwise suggest
feat/<slug>,fix/<slug>,docs/<slug>, orchore/<slug>. - Never suggest
HEAD.
- Keep the current non-
- Use an imperative title with a conventional prefix when useful.
- Start the description with
This pull request adds ...,fixes ...,improves ..., orupdates .... - Explain the motivation, complete behavioral change, and compatibility considerations. Do not list tests unless the user asks.
- Normalize GitHub references:
#123for this repository andowner/repo#123for another repository. Remove Markdown-linked issue/PR labels, bare issue/PR URLs, local paths, Codex citations, and app directives. - Return the following block in English.
Output format
# Pull Request Draft
## Branch name suggestion
git checkout -b <kebab-case branch>
## Title
<single-line imperative title>
## Description
<copy-ready description beginning with "This pull request ...">
Keep the block tight and avoid repeating the same information in multiple sections.
04 同仓库更多 Skills
maintainer-review
Assess an openai-guardrails-python GitHub issue or pull request as a maintainer. Use to verify the claimed need and practical impact, compare supported alternatives or competing approaches, separate code quality from repository readiness, recommend the maintainer action, and draft a copy-ready comment when evidence, changes, or closure should be requested.
implementation-strategy
Choose compatibility-aware scope for runtime and API changes in openai-guardrails-python. Use before initial implementation and each review-feedback batch to decide whether to patch, reset the design, preserve compatibility, or reject unsupported cases.
implementation-kickoff
Start and carry an explicitly invoked openai-guardrails-python implementation through a fresh isolated worktree and a local PR-ready handoff. Fetch the latest origin/main, keep task changes uncommitted, replay them onto the latest main before final review, run applicable verification and $implementation-final-review, use $pr-draft-summary to generate the complete PR draft and branch name, then create one clean local commit with takeover provenance when applicable. Use only when the user explicitly invokes this skill; never push, open a PR, or mutate GitHub.
implementation-final-review
Perform the repository's risk-tiered independent final review before implementation completion. Use only when explicitly invoked or when repository instructions require it after behavior-impacting implementation work; audit the complete task diff, supported contracts, lifecycle and security boundaries, complexity, and tests before final verification.
final-release-review
Review an openai-guardrails-python release plan or final release candidate against the previous remote tag, determine the compatible release type, audit runtime and packaging risk, inspect current CI, and produce an English ship-or-block report. Use for pre-release readiness checks, not ordinary PR review or implementation.
code-change-verification
Run the mandatory verification stack when changes affect runtime code, tests, examples, or build and test behavior in the OpenAI Guardrails Python repository.