SkillVaultskills Browse all 1,000+ skills

Delivery · Version 1.1.0 · Reviewed 2026-08-02

Pull Request Publisher

Make a defensible decision about change-set readiness check and branch and commit publication with evidence, explicit trade-offs, and a verification plan.

4 method steps 5 documented failure modes 5 diagnostic checks 7 quality gates

Publishes a review-ready pull request from an existing change set by checking worktree safety, branch intent, validation evidence, commit quality, remote state, and repository-specific PR requirements.

₹199 one-time

Get this skill archive

Install in your AI coding tool

SkillVault packages this skill in the open Agent Skills format for five leading coding tools.

What this skill helps you do

  • Change-set readiness check
  • Branch and commit publication
  • Pull request creation

How Pull Request Publisher works

You provide

The intended change, worktree, target remote and base, plus repository contribution rules

It inspects

Merge-base diff, unrelated files, branch divergence, required checks, and rollout risk

It decides

A focused commit and review-ready pull request whose claims match the published revision

You verify

Remote head, pull request metadata, and reported validation evidence agree exactly

What it checks first

Pull Request Publisher publishes a review-ready pull request from an existing change set by checking worktree safety, branch intent, validation evidence, commit quality, remote state, and repository-specific PR requirements. Use it when the work involves Change-set readiness check, Branch and commit publication, Pull request creation.

  1. The authoritative merge base and complete base comparison, because the working tree alone omits committed branch changes.
  2. Staged, unstaged, untracked, generated, and unrelated changes that would make a supposedly focused pull request misleading.
  3. Repository pull request templates, contribution rules, branch protections, required checks, and commit conventions.
  4. Validation evidence tied to the changed behavior, including exact commands, failures, skipped checks, and environmental limits.
  5. Remote tracking and divergence, especially whether publishing would overwrite shared history or target the wrong repository.

Failure modes it recognizes

  • A pull request is created from the wrong base branch, mixing old commits into the review and hiding the intended change boundary.
  • A broad stage-all operation includes local configuration, generated output, credentials, or unrelated edits in the published branch.
  • The description claims tests passed from a previous revision even though later commits changed the behavior those tests covered.
  • A force push rewrites a branch another reviewer or automation already consumed, invalidating comments and external references.
  • The pull request title and body restate filenames while omitting user impact, rollout order, migration risk, and rollback constraints.

Answers it will reject

  • Publishing first and promising to clean the branch later; reviewers form conclusions from the initial noisy comparison.
  • Using a green build badge as proof that every material change was tested without mapping checks to the touched boundaries.
  • Inventing issue links, test outcomes, screenshots, or deployment status to make the pull request look complete.
  • Silently changing the requested base, remote, reviewers, labels, or branch history because an automation default selected them.

Decision rules it applies

  • Stop before staging when intended and unrelated changes cannot be separated confidently from the available evidence.
  • Use non-destructive push behavior by default; rewriting published history requires explicit scope and confirmation.
  • Create the pull request only after its title, description, commit set, and validation claims match the final remote diff.
  • Treat migration, security, public contract, and deployment changes as requiring explicit risk and rollback sections.

Evidence it asks for

  • Inspect status, staged and unstaged diffs, untracked files, upstream divergence, and the full merge-base comparison.
  • Run the smallest repository-defined validation commands that cross every changed contract and preserve their exact outcomes.
  • Resolve the pushed remote ref and compare its commit identifier with the local branch before creating the pull request.
  • Read the created pull request back from the hosting provider to verify base, head, title, body, and URL.

The method inside

  1. Establish what is actually true about change-set readiness check from the supplied evidence, and mark what is missing.
  2. Identify the mechanism behind branch and commit publication rather than restating the symptom.
  3. Choose the smallest defensible change for pull request creation, weighing impact, confidence, effort, and reversibility.
  4. Record consequences, rollback, and open questions

Deliverables

  • Change-set readiness check assessment
  • Branch and commit publication decision and action plan
  • Pull request creation verification checklist

Evidence requirements

  • Functional and quality requirements
  • Scale, latency, consistency, cost, and compliance constraints
  • Current topology and alternatives considered

Quality gates

  • Every material claim traces to supplied evidence or is labeled as a hypothesis.
  • The response follows the declared deliverable contract.
  • No execution, access, measurement, or verification is invented.
  • Secrets and personal data are redacted rather than repeated.
  • The user receives a concrete independent verification step.
  • The relevant failure modes in this domain were considered rather than only the reported symptom.
  • No listed anti-pattern was recommended as a solution.

Example task

Input

Publish my current changes as a pull request without losing unrelated work, and use the repository template plus the validation results already available.

Expected output

First separate intended changes from unrelated work and stop if the branch or remote state is ambiguous. Then run the smallest repository-defined validation, create focused commits, push without rewriting shared history, and publish a PR whose claims match the actual diff and test evidence...

Boundaries and compatibility

Ideal for

  • Change-set readiness check: produce a decision or artifact grounded in supplied evidence.
  • Branch and commit publication: produce a decision or artifact grounded in supplied evidence.
  • Pull request creation: produce a decision or artifact grounded in supplied evidence.

Out of scope

  • Producing a generic reference architecture without requirements
  • Hiding material trade-offs behind best-practice language

Agent compatibility

  • GitHub Copilot Agent Skills
  • Cursor Agent Skills
  • Claude Code Skills
  • OpenAI Codex Skills
  • JetBrains Junie Skills

Tool policy: Advisory by default. No tools are assumed. If the host provides tools, use read-only evidence gathering unless the user explicitly approves a scoped write or execution action.