What does this skill do, and when should you use it?
This skill updates the existing pull request for the current session. It checks whether the pull request contains commits missing from the current branch, then pulls those changes and resolves merge conflicts when necessary. It runs compile and hygiene tasks, commits any remaining uncommitted changes, and revises the pull request title and description when the changes are significant. It then updates the pull request with the new commits and information.
Reads the pull request information appended to the prompt; checks for commits present in the pull request but absent from the current branch; pulls incoming changes and resolves merge conflicts; runs compile and hygiene tasks and fixes errors; invokes the “/commit” skill for uncommitted changes; updates the pull request title and description when changes are significant; and updates the pull request with the new commits and information.
- A maintainer is continuing work on an existing pull request and must incorporate incoming commits first.
- A developer has made follow-up changes and needs compilation and hygiene checks before publishing them.
- A pull request has changed substantially and its title and description need to be kept accurate.
How do you install this skill?
- Obtain explicit user confirmation before pulling, merging, committing, or pushing external changes, and preserve a recoverable branch or commit point.
- Document exact commands, stop conditions for unresolved conflicts, failure diagnostics, and credential or sensitive-data handling.
- Define behavior when no PR exists, no incoming changes exist, GitHub is unavailable, or the remote repository cannot be reached.
- Shell / CLI
- Network access
- Local filesystem
Git/commit skill
The skill is located at src/vs/sessions/skills/update-pr/SKILL.md in the microsoft/vscode repository. The supplied README documents the repository and the bundled collection, but does not provide a standalone installation command for this skill; no additional installation steps can be verified from the source.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/microsoft/vscode.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/src/vs/sessions/skills/update-pr" ~/.claude/skills/
rm -rf "$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:
- Update the existing pull request for the current session and push my new changes.
In a client that supports the skill, use a request such as “Update the existing pull request for the current session and push my new changes.” The prompt must contain the relevant pull request information; the source does not document a specific trigger syntax.
What are this skill's strengths and limitations?
- Covers synchronization, conflict handling, compilation, hygiene checks, committing, and pull request metadata updates in one workflow.
- Targets the existing pull request associated with the current session, which suits incremental development.
- Uses a minimal SKILL.md with standard name and description frontmatter.
- The source does not specify the compile commands, hygiene tasks, or conflict-resolution procedure.
- It depends on an external “/commit” skill whose implementation and installation are not provided.
- No test suite, platform support matrix, or failure-recovery details are documented.
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 |
|---|---|---|---|---|
| Existing Pull Request Updater this page ✓ Microsoft · Official | 32 · Not recommended | ★ 194k | 11d ago | MIT |
| Codex Pull Request Editor ✓ OpenAI · Official | 50 · Use with care | ★ 129k | 11d ago | Apache-2.0 |
| Session Pull Request Assistant ✓ Microsoft · Official | 43 · Not recommended | ★ 194k | 11d ago | MIT |
| Yeet GitHub PR Flow ✓ OpenAI · Official | 41 · Not recommended | ★ 28k | 4mo ago | — |
| CODEOWNERS PR Splitter ✓ NVIDIA · Official | 57 · Use with care | ★ 3.6k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill explicitly directs the agent to pull incoming changes, resolve conflicts, commit, and push updates to an existing PR, creating meaningful external state changes. It lacks staged user confirmation, credential and data-flow disclosure, rollback guidance, and least-privilege scoping, so points are deducted. No red-line risk in the supplied material is established.
The sequence is internally coherent, and the repository context provides compile, hygiene, and commit-related capabilities. However, the actual commands, conflict-resolution policy, failure diagnostics, and abnormal-input feedback are unspecified. This is a static review, so the calibration cap keeps the score at or below 10.
The name and description clearly target updating an existing PR for the current session. Boundaries are missing for no existing PR, no incoming changes, unresolvable conflicts, non-GitHub environments, and other unsupported cases; inputs, outputs, Chinese-language support, and network fit are also unspecified.
The skill has a concise name, trigger description, and linear procedure, while repository context supplies MIT licensing and official Microsoft provenance. The skill itself lacks installation and dependency notes, stable parameters, examples, FAQs, known limitations, version/change history, and an explicit maintenance path, so points are deducted.
The procedure covers synchronization, compilation and hygiene checks, committing, and PR updates, so the core outcome is plausible. However, compile and hygiene tasks, significance thresholds, completion criteria, and output verification are undefined; static evidence cannot establish directly usable results, so points are deducted.
The skill text is auditable, package.json exposes compile and hygiene scripts, and some CI provides repository-level execution evidence. The supplied CI does not cover this skill's key path, and there are no dedicated tests, representative outputs, or independent reproduction evidence, so points are deducted.
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 →