Finishing a Development Branch
obra/superpowers
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.
Create a pull request, optionally for the issue provided as argument
$ npx skills add mengxi-ream/read-frog --skill create-pr -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install mengxi-ream/read-frog create-pr --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/create-pr .claude/skills/create-pr && rm -rf skills-srcUse ~/.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/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .claude/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-prType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add mengxi-ream/read-frog --skill create-pr -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install mengxi-ream/read-frog create-pr --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/create-pr .agents/skills/create-pr && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .agents/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add mengxi-ream/read-frog --skill create-pr -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install mengxi-ream/read-frog create-pr --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/create-pr .cursor/skills/create-pr && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .cursor/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/mengxi-ream/read-frog.git --path .agents/skills/create-pr--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add mengxi-ream/read-frog --skill create-pr -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install mengxi-ream/read-frog create-pr --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/create-pr .gemini/skills/create-pr && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .gemini/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install mengxi-ream/read-frog create-prInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add mengxi-ream/read-frog --skill create-pr -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/create-pr .github/skills/create-pr && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .github/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add mengxi-ream/read-frog --skill create-pr -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install mengxi-ream/read-frog create-pr --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mengxi-ream/read-frog.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/create-pr .opencode/skills/create-pr && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "create-pr" agent skill from https://github.com/mengxi-ream/read-frog/tree/main/.agents/skills/create-pr into .opencode/skills/create-pr/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "create-pr", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
create-prCreate a pull request, optionally for the issue provided as argument
Create PR is an agent skill from mengxi-ream/read-frog. Create a pull request, optionally for the issue provided as argument
Its SKILL.md is about 2.4k 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. It works with Chrome Extensions and Git. The repository describes itself as: 🐸 Read Frog - Language Learning & Translate | 🐸 陪读蛙 - 语言学习与翻译. The licence is GPL-3.0.
11 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 8a29c04. It shows what the files ask for, not the result of running them.
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.
Shell commands in SKILL.md call:
gitghpnpmFrom the folder's file list and the shell code blocks in SKILL.md.
Hosts in commands or code, which the agent is likely to contact:
github.comFrom URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Create PR loads about 2.4k tokens when it runs. Until then it costs about 20 tokens; SKILL.md has 1,309 words of instructions outside code blocks.
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.
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.
The full file from mengxi-ream/read-frog at commit 8a29c04, republished under its GPL-3.0 licence (© mengxi-ream). 1,309 words, ~2,429 tokens.
.claude/skills/create-pr/SKILL.md (or your agent's skills folder).Create a pull request. If an issue number is provided in $ARGUMENTS, use it. If no issue is provided, continue without requiring one.
Follow these steps:
Check current git status and branch information
git status and git branch to understand the current stateCreate branch if needed
Review commit history and code differences
git log and git diff main...HEAD to understand all changes from the main branchRecord a demo for user-visible frontend changes
Require a demo when the diff changes user-visible behavior: injected page UI, popup / options / side-panel UI, translation output shape, interactions, routing inside extension pages, or loading and error states.
Do not require one for implementation-only refactors, types, tests, build or CI config, or anything with no visible effect.
When a demo is required, invoke the record-extension-demo skill. If it is unavailable, follow the prompt below with the browser tooling at hand; if faithful recording is not possible, report the limitation instead of fabricating evidence.
Record a concise PR demo for <feature> on the current branch against <base-branch>. Inspect the diff and the repository instructions first. Turn the behavior into captioned scenes with setup, action, assertion, and end state. Build `.output/chrome-mv3` and drive it in a fresh Chrome profile with the whole config forced from the service worker and `targetCode = 'cmn'`; prefer `microsoft-translate-default` so no API key can reach a frame. Default desktop capture to 1280x800, but choose another explicit viewport when the surface requires it (the popup is small). Gate every transition on a selector or a translated-node count rather than a fixed sleep; after the assertion succeeds, hold the state about two to three seconds. Add caption pills and click highlights marked `notranslate` without modifying extension source. For bug fixes or visual regressions, rebuild the base commit in a separate worktree and capture the same scenario on both builds; for static visual changes, include paired screenshots. Keep automation and media in the scratchpad, remove install and onboarding prelude, and measure any crop from a full-resolution frame. Visually QA the result and return the absolute MP4 path, optional GIF path, scene captions, assertion results, baseline/candidate SHAs when compared, and the functional verification summary. Exclude API keys, tokens, personal tabs, and unrelated notifications. If the environment cannot reproduce the behavior faithfully, stop and report the limitation.Keep the recording in the session scratchpad. Never commit a video or GIF.
Treat the recording as verification evidence: do not tick Verified through manual testing unless both the state assertions and the visual QA passed.
If no demo is required, write Not applicable — no user-visible change. in the PR template's ## Screenshots section.
Upload the demo to GitHub attachments — only when step 4 produced a recording
Skip this step entirely when step 4 decided no demo was required, or when it reported that the environment could not reproduce the behavior faithfully. There is no file to upload, and the ## Screenshots section already says so. Go straight to step 6.
Use gh-image so the PR embeds a repository-scoped user-attachments video instead of a third-party asset host.
Check before uploading:
gh auth status
gh extension list | grep 'drogers0/gh-image'
gh image --versionRequire v1.1.0 or newer. If it is missing or outdated, tell the user to run gh extension install drogers0/gh-image or gh extension upgrade gh-image and wait. Do not install it on their behalf.
Never run gh image extract-token, pass --token, or handle a GitHub user_session cookie. The normal gh token is enough for a repository the current account can push to.
Confirm the finished recording shows no API keys, tokens, personal tabs, or unrelated information. This repository is public and uploads cannot be undone.
Upload the verified MP4 and capture the bare URL printed to stdout. Upload a GIF only when the destination needs one:
gh image "/absolute/path/to/demo.mp4" --repo mengxi-ream/read-frogAccept the result only when it is one bare https://github.com/user-attachments/assets/... URL. Put it inside stable markers in the ## Screenshots section so a rerun replaces the section instead of appending a second video:
<!-- read-frog-pr-demo:start -->
https://github.com/user-attachments/assets/...
- Scene 1: <caption and assertion>
- Verification: <DOM/storage check and result>
<!-- read-frog-pr-demo:end -->GitHub renders the bare video URL as an inline player. When updating an existing PR, replace exactly one marked region and leave the rest of the body unchanged. Do not re-upload a file merely because the body update failed.
Add changeset record if necessary
Manually add a file in .changeset/ directory following changeset convention
The package name is always @read-frog/extension. This is a single-package workspace and changesets matches the header against the name in the root package.json; a bare read-frog is not a workspace package and makes the release plan fail.
File shape — fill in the bump level and the summary, leave the package name as written:
---
"@read-frog/extension": patch
---
fix(subtitles): follow YouTube's own default caption trackChangeset summary should use conventional commit style and should match the descriptive PR title
Versioning rules:
patch (0.0.x) — Users barely noticeminor (0.x.0) — Users can clearly perceive "something new"major (x.0.0) — Users need to pay attention / adapt. Almost never use. Must ask user for explicit approval before using.Verify the changeset before committing: pnpm exec changeset status must list the expected bump under @read-frog/extension. A non-zero exit means the changeset is malformed and the release plan will not generate.
Ensure all changes are committed
Push the branch to remote
git push -u origin <branch-name> if neededCreate PR with GitHub CLI
gh pr create with:.github/PULL_REQUEST_TEMPLATE.mdCloses #<issue-number>user-attachments URL and the demonstrated scenes in the ## Screenshots section, inside the read-frog-pr-demo markers — or the step 4 sentence saying why there is noneVerify the demo embed — only for a PR that has one
Skip when no demo was required. Otherwise confirm the attachment survived the body without printing the whole body:
gh pr view <pr> --repo mengxi-ream/read-frog --json body -q .body | grep -c 'github.com/user-attachments/'Expect at least one match. If the embed is missing, fix the PR body; do not upload the video again.
not applicable, or the exact recording/upload limitation.Follow these commit types:
feat: New featuresfix: Bug fixesdocs: Documentation changesstyle: Code style changesrefactor: Code refactoringtest: Adding or updating testschore: Maintenance tasksbuild: Build system changesci: CI/CD changesperf: Performance improvementsrevert: Reverting previous commitsi18n: Internationalization changesai: AI-related featuresFormat: type(scope): description
.github/PULL_REQUEST_TEMPLATE.mdCLAUDE.md© mengxi-ream, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Just SKILL.md in .agents/skills/create-pr of mengxi-ream/read-frog.
Open the folder on GitHubat commit 8a29c04
Create PR 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Create PR this skillmengxi-ream/read-frog | 10k | — | ~2.4k | Automated safety check: Pass | GPL-3.0 | |
| Finishing a Development Branchobra/superpowers | 296k | 5 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Understand Diff AnalysisEgonex-AI/Understand-Anything | 85k | 1 repos | ~1.4k | Automated safety check: Pass | MIT | |
| Contributor-First PR MergeHKUDS/OpenHarness | 16k | 1 repos | ~847 | Automated safety check: Pass | MIT | |
| Open Code Review CLIalibaba/open-code-review | 44k | — | ~3.1k | Automated safety check: Pass | Apache-2.0 | |
| Create Pull Requestcline/cline | 70k | 1 repos | ~1.6k | Automated safety check: Pass | Apache-2.0 |
obra/superpowers
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.
Egonex-AI/Understand-Anything
Reads your git changes or a pull request against a prebuilt knowledge graph of the project to explain what changed, which components are affected and what is risky.
HKUDS/OpenHarness
Merges external GitHub pull requests while keeping the original author credited, and fixes conflicts after the merge instead of rewriting the contribution.
alibaba/open-code-review
Runs the ocr command-line tool to review Git changes, a commit or a branch comparison with an AI model, returning line-level comments and optionally applying fixes.
cline/cline
Opens a GitHub pull request from your current branch with the gh CLI, after reviewing the commits and diff and gathering the details the PR needs.
openinterpreter/openinterpreter
Rewrites the title and body of one or more pull requests with gh, leading with why the change was made, then what changed, and describing only the net result.
mengxi-ream/read-frog
Adversarial code review using cross-model approach. An agent skill from mengxi-ream/read-frog.
mengxi-ream/read-frog
Record polished, evidence-backed demos of the Read Frog extension as MP4 (and optional GIF) by driving real Chrome with the built extension, following captioned scene scripts, asserting extension…
mengxi-ream/read-frog
Debug the built Read Frog extension in real Chrome. An agent skill from mengxi-ream/read-frog.
mengxi-ream/read-frog
Utilities for browser extensions - proxy services for cross-context RPC, type-safe messaging, URL match patterns, fake browser for testing, job scheduling, and shadow DOM isolation.
mengxi-ream/read-frog
Build browser extensions with WXT - a next-gen framework with file-based entrypoints, Vite builds, auto-imports, typed storage with migrations, content script UIs, and multi-browser MV2/MV3 support.
mengxi-ream/read-frog
A skill your agent uses when writing or reviewing database/config migration scripts or migration registries.
Works with
Categories
Create a pull request, optionally for the issue provided as argument. Create PR is an agent skill from mengxi-ream/read-frog.
Create PR fits situations like: tasks that involve Pull requests.
Run `npx skills add mengxi-ream/read-frog --skill create-pr -a claude-code`. Or copy the skill folder (.agents/skills/create-pr in mengxi-ream/read-frog) into .claude/skills/create-pr in your project. Claude Code loads it when a task matches its description.
Run `npx skills add mengxi-ream/read-frog --skill create-pr -a codex`. Or copy the skill folder (.agents/skills/create-pr in mengxi-ream/read-frog) into .agents/skills/create-pr in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mengxi-ream/read-frog --skill create-pr -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, .gemini/skills/create-pr, .github/skills/create-pr and .opencode/skills/create-pr in your project.
Going by SKILL.md and its folder, Create PR needs the command-line tools its instructions call (git, gh and pnpm).
SKILL.md names 1 domain. In commands or code: github.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.
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.
Create PR is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.4k tokens (SKILL.md is roughly 9.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Create PR: Finishing a Development Branch (obra/superpowers, 296k stars), Understand Diff Analysis (Egonex-AI/Understand-Anything, 85k stars), Contributor-First PR Merge (HKUDS/OpenHarness, 16k stars) and Open Code Review CLI (alibaba/open-code-review, 44k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
mengxi-ream (a GitHub user) maintains it in mengxi-ream/read-frog, which has 9,987 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on October 7, 2026.
Source: mengxi-ream/read-frog on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.