What does this skill do, and when should you use it?
Yeet is a GitHub CLI-oriented workflow skill for GitHub pull requests. It runs only when the user explicitly asks for staging, committing, pushing, and opening a pull request as one flow. It checks GitHub CLI availability and authentication, handles branches and commits, runs checks, discovers PR templates, and creates or updates the relevant PR. The skill is part of the deprecated openai/skills catalog, whose README directs users to the OpenAI Plugins repository for current examples.
Checks gh --version and gh auth status; creates a description-named branch when starting from the default branch; runs git status -sb, git add -A, and a description-named commit; runs checks when needed; pushes the current branch with tracking; discovers PR templates from the repository root; checks whether the current branch already has a PR; updates an existing PR or creates a new draft PR; and edits the PR title and body to match the actual net diff.
- A developer has finished a GitHub change and explicitly wants the current workspace taken through PR creation.
- A team member wants the same description reused for the branch, commit, and PR subject.
- A maintainer needs the repository's active PR template read before a PR is created.
- An existing PR author wants subsequent changes applied to the current branch's PR instead of opening another PR.
How do you install this skill?
- Review git status and the full diff before execution; git add -A may include unintended files or sensitive information.
- Push, pull, and PR creation change external repository state, but the skill does not define stepwise confirmation or rollback procedures.
- GitHub CLI, GitHub authentication, and network reachability are hard dependencies and may affect mainland-China usability.
- The repository is marked deprecated, so ongoing maintenance and compatibility should be confirmed separately.
- Shell / CLI
- Network access
- Local filesystem
GitGitHub CLI (gh)authenticated GitHub CLI session
The skill is located at skills/.curated/yeet/. In Codex, install it with $skill-installer yeet; restart Codex afterward so it can load the skill. The README does not document another installation method or a standalone license.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/.curated/yeet" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Use GitHub CLI to stage and commit the current changes, push them, and open a draft pull request.
Trigger it only with an explicit request for the complete flow, such as: “Use GitHub CLI to stage and commit the current changes, push them, and open a draft pull request.” gh must be installed and authenticated through gh auth login; if a prerequisite check fails, the workflow asks for remediation and stops.
What are this skill's strengths and limitations?
- Covers the workflow from Git status confirmation through PR creation or update.
- Discovers and reads an applicable PR template from the repository root.
- Updates an existing PR in place and preserves its draft or ready-for-review state.
- Requires the final PR title and body to reflect the actual net diff.
- It is scoped to an explicitly requested end-to-end GitHub PR flow, not ordinary commit or push assistance.
- It depends on GitHub CLI, Git, and an authenticated GitHub session.
- It stops before PR creation when multiple templates are found and asks the user to choose.
- The source does not provide a test suite, platform coverage, or standalone license evidence.
- The repository README states that the repository is deprecated.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Yeet GitHub PR Flow this page ✓ OpenAI · Official | 41 · Not recommended | ★ 28k | 4mo ago | — |
| Codex Pull Request Editor ✓ OpenAI · Official | 50 · Use with care | ★ 129k | 11d ago | Apache-2.0 |
| CODEOWNERS PR Splitter ✓ NVIDIA · Official | 57 · Use with care | ★ 3.6k | 11d ago | Apache-2.0 |
| Ship a PR (e2e repo PR delivery workflow) | 59 · Recommended | ★ 10k | 11d ago | Apache-2.0 |
| RenderCV PR Reviewer | 55 · Use with care | ★ 18k | 7mo ago | MIT |
How did FollowSkills review this skill?
The skill requires an explicit user request and checks gh authentication, providing basic scope control; however, it unconditionally performs git add -A, commit, push, and PR creation without stepwise confirmation, sensitive-change review, rollback guidance, or adequate disclosure of external effects from pushing and dependency installation, so substantial points are deducted.
The workflow is generally coherent and includes prerequisites, template discovery, and existing-PR checks; however, there is no test suite or key-path reproduction evidence, the checks to run are unspecified, dependency installation can alter the environment, and pulling from master after a push failure may introduce unexpected changes. Static scoring is therefore limited.
The trigger and intended GitHub workflow are clear, and the skill is explicitly restricted to a user-requested end-to-end flow; however, non-fit cases, confirmation boundaries, and behavior outside GitHub or gh environments are not well defined. Dependence on GitHub CLI and GitHub network services may also reduce reachability for mainland-China users.
The documentation is readable and layered, covering prerequisites, naming, templates, workflow, titles, and PR-body rules, and it notes that the repository is deprecated; however, the skill's license, version, changelog, maintenance owner, and update path are not clear, and installation/resource documentation is incomplete.
The text plausibly covers staging, committing, pushing, and creating or updating a PR, including template handling and existing-PR preservation; however, no representative output, successful result, or behavioral evidence is provided. Automatic staging and retry behavior may require substantial manual review, limiting the static effectiveness score.
The skill provides concrete commands, flags, and branching conditions, making the intended process auditable; however, there is no committed test suite, CI evidence, third-party execution evidence, or corroborating source material, so higher verifiability cannot be justified.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →