gh-create-pr
Create or update GitHub pull requests using the repository-required workflow and template compliance
Install / Use
npx skills add CherryHQ/cherry-studio --skill gh-create-prInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
AutomationSupported Platforms
Our assessment of gh-create-pr
gh-create-pr scores 87/100 on our quality scale, 353rd of 1,111 Automation skills we index (top 32%).
Its SKILL.md is 7.1 KB long, split into 7 sections with 1 code example: a thorough specification that gives an agent plenty to work with.
With 52,134 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated today, so gh-create-pr is actively maintained.
- It is released under AGPL-3.0, a copyleft license: you can use it, but modified versions you distribute must carry the same license.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
gh-create-pr compared with similar skills
All 4 of these similar skills score higher than gh-create-pr; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| gh-create-pr (this skill)by CherryHQ | 87 | 52.1k | today | SKILL.md |
| Agent-Reachby Panniantong | 100 | 85.4k | 9d ago | CLAUDE.md |
| rufloby ruvnet | 100 | 73.2k | today | CLAUDE.md |
| Scraplingby D4Vinci | 100 | 83.5k | today | MCP Server |
| algorithmic-artby anthropics | 100 | 177.9k | 2d ago | SKILL.md |
Frequently asked questions
- How do I install gh-create-pr?
- Run
npx skills add CherryHQ/cherry-studio --skill gh-create-pr. The install tabs above show the steps for each supported agent. - Which AI agents does gh-create-pr work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is gh-create-pr safe to use?
- It is AGPL-3.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is gh-create-pr still maintained?
- The repository was last updated today, so gh-create-pr is actively maintained.
Skill content
View source on GitHubname: gh-create-pr
description: Create or update GitHub pull requests using the repository-required workflow and template compliance. Use when asked to create/open/update a PR so the assistant reads .github/pull_request_template.md, fills every template section, preserves markdown structure exactly, and marks missing data as N/A or None instead of skipping sections.
GitHub PR Creation
Workflow
- Read
.github/pull_request_template.mdbefore drafting the PR body. - Collect PR context from the current branch (base/head, scope, linked issues, testing status, breaking changes, release note content).
- Check if the current branch has been pushed to remote. If not, push it first:
- Default remote is
origin, but ask the user if they want to use a different remote.
git push -u <remote> <head-branch> - Default remote is
- Determine the base branch:
- Inspect the head branch before applying any default. If the head is
release/v<version>, stop: never open a pull request from a release branch, especially not tomain. Put an isolated fix on a topic branch and target the release branch, or let Post Release create the metadata sync branch. - Before defaulting an arbitrary topic branch to
main, inspect its merge base and upstream. A product-code fix that started fromrelease/v<version>must stop and be recreated frommainas a hotfix; only a release-only repair or explicit backport recovery topic may target that exact release branch. - A
backport/v<version>/pr-<number>head must target the matchingrelease/v<version>base and its body must contain<!-- release-backport-source-pr: <number> -->on its own line for the exact source hotfix PR. Arelease-sync/v<version>head must targetmain, use the exact titlechore(release): sync v<version> metadata, retain therelease-metadata-boundary: v<version>body marker, and be squash-merged. - For official repo(CherryHQ/cherry-studio) as
origin: default base ismainfromorigin, but allow the user to explicitly indicate a base branch. mainis the active development line, including hotfix PRs. Do not target an old maintenance branch unless the user explicitly requests it.- Only classify a PR as a release hotfix when the user explicitly says it must be included in the active draft release. Use the title
hotfix: <description>orhotfix(<kebab-case-scope>): <description>with a lowercase alphanumeric kebab-case scope, exactly one space after the colon, and a non-empty description. The title grammar synchronizes thehotfixlabel automatically. Merging opens a separate backport PR only when exactly one draft semantic-version release has a matching activerelease/v<version>branch; otherwise automation stops without guessing a target. - If the hotfix is user-facing, provide one release-note line in each language so automation can update the active draft and stable release history. Put this exact structure inside the template's existing
release-notefence; do not include bullet prefixes:<!--LANG:en--> [Component] English description. <!--LANG:zh-CN--> [组件] 中文说明。 <!--LANG:END--> - If the hotfix has no user-facing release note, keep
NONEin the template'srelease-notefence. The code is still backported, but release metadata is unchanged. A provided bilingual block must satisfy the exact structure above. - For fork repo as
origin: check available remotes withgit remote -v, default base may beupstream/mainor another remote. Always assume that user wants to merge head to CherryHQ/cherry-studio/main, unless the user explicitly indicates a base branch. - Ask the user to confirm the base branch if it's not the default.
- Inspect the head branch before applying any default. If the head is
- Create a temp file and write the PR body using a single Bash heredoc
(avoids
mktemp+Writetool path-mismatch on Windows):
Fill content using the template structure exactly (keep section order, headings, checkbox formatting). If not applicable, writepr_body_file="/tmp/gh-pr-body-$(date +%s).md" cat > "$pr_body_file" <<'EOF' ...filled template body... EOFN/AorNone. - Preview the temp file content via Bash
cat "$pr_body_file"(theReadtool can fail on/tmp/...paths on Windows). Show the file path and ask for explicit confirmation before creating. Skip if the user explicitly waives preview (automation workflows). - After confirmation, create the PR:
For a release hotfix, use an exact classifier-compatible title so the workflow synchronizes the label and separately validates any provided bilingual note:gh pr create --base <base> --head <head> --title "<title>" --body-file "$pr_body_file"
After creation, verify that the classifier addedgh pr create --base main --head <head> --title "hotfix: <description>" --body-file "$pr_body_file"hotfix; if it did not, fix the title instead of adding the label manually. Also fix any malformed optional release-note block reported by the workflow. - Clean up the temp file:
rm -f "$pr_body_file" - Report the created PR URL and summarize title/base/head and any required follow-up.
Constraints
-
Never skip template sections.
-
Never rewrite the template format.
-
Keep content concise and specific to the current change set.
-
PR title and body must be written in English.
-
Never use a
hotfixtitle or label for an ordinary bug fix. It opts the PR into an automatic backport PR after merge when one matching draft release is active. -
A release hotfix may use
NONEwhen it has no user-facing release note. Never provide a partial, single-language, or otherwise malformed bilingual note; the backport workflow fails closed on malformed content. -
Never default a
release/v<version>head tomain; a release branch is not a pull request source branch. -
Never create the PR before showing the full final body to the user, unless they explicitly waive the preview or confirmation.
-
Never rely on command permission prompts as PR body preview.
-
Release note & Documentation checkbox — both are driven by whether the change is user-facing. Use the table below:
| Change type | Release note | Docs
[x]| |---|---|---| | New user-facing feature / setting / UI | Describe the change | ✅ | | Bug fix visible to users | Describe the fix | ✅ if behavior changed | | Behavior change / default value change | Describe +action required| ✅ | | Security fix in a user-facing dependency | Describe the fix | ✅ if usage changed | | CI / GitHub Actions changes |NONE| ❌ | | Internal refactoring (user cannot tell) |NONE| ❌ | | Dev / build tooling changes |NONE| ❌ | | Dev-only dependency bump |NONE| ❌ | | Test-only / code style changes |NONE| ❌ |
Command Pattern
# read template
cat .github/pull_request_template.md
# show this full Markdown body in chat first
pr_body_file="/tmp/gh-pr-body-$(date +%s).md"
cat > "$pr_body_file" <<'EOF'
...filled template body...
EOF
cat "$pr_body_file"
# run only after explicit user confirmation
gh pr create --base <base> --head <head> --title "<title>" --body-file "$pr_body_file"
rm -f "$pr_body_file"
Related Skills
Agent-Reach
85.4kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
ruflo
73.2k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
Scrapling
83.5k🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
