happier-pr-steward

Shepherd Happier pull requests through review, refinement, porting, and merge assessment.

1.6k|141|Updated Dec 16, 2025
One-click install
npx skills add https://github.com/happier-dev/happier --skill happier-pr-steward
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: happier-pr-steward
Source: https://github.com/happier-dev/happier/tree/main/skills/happier-pr-steward
Command: npx skills add https://github.com/happier-dev/happier --skill happier-pr-steward

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Managing a pull request end-to-end involves scattered judgment calls: verifying whether the PR actually solves the right problem, adjudicating reviewer and bot feedback against evidence, keeping CI and review state tied to the current head, and porting accepted changes across version lines. This Skill orchestrates that entire lifecycle with explicit human gates so nothing is merged, pushed, or claimed without evidence.

Core Features & Use Cases

  • Evidence-based PR review: Reconstructs the PR's intent, identifies the canonical code owner, detects duplicate or split-brain logic, and produces a merge verdict (mergeable, mergeable after refinements, or not recommended).
  • Feedback adjudication: Reproduces each review finding from current source, classifies it as confirmed, stale, invalid, or not applicable, and applies only confirmed corrections with TDD.
  • Cross-version porting: Carries a 0.2-line PR and every accepted follow-up into the 0.3 line with preserved intent and per-commit co-author attribution.
  • Use Case: A maintainer asks whether a contributor's PR is safe to merge; the Skill fetches the live head, audits the diff, posts refinements, monitors CodeRabbit and human reviewers, fixes confirmed findings, ports them to the 0.3 branch, and reports final SHAs and residual risk.

Quick Start

Use the happier-pr-steward skill to analyze pull request number 123 in the happier repository, propose evidence-backed refinements, and shepherd the approved changes through review.

Frequently Asked Questions about happier-pr-steward

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review whether a GitHub pull request is mergeable?

The Skill fetches the live PR head, diff, reviews, and checks, then reconstructs the intended problem and compares the patch against the canonical owner. It reports a merge verdict of mergeable, mergeable after named refinements, or not recommended, backed by evidence.

How to handle reviewer and bot feedback on a pull request?

Each finding is reproduced against current source, then classified as confirmed, already fixed, invalid, stale, or not applicable. Only confirmed in-scope corrections are implemented with tests, and rejected suggestions receive concise factual reasons.

Can it port changes from one release branch to another?

Yes. When a PR belongs to the 0.2 line, it invokes the happier-port-0-2-to-0-3 skill to carry the full PR intent plus every accepted follow-up into the 0.3 line, then validates and commits the destination change with preserved attribution.

Does the skill merge or push to GitHub automatically?

No. Merges, pushes, comments, and reviewer requests require exact authorization or an explicit bounded standing grant. Under exact authorization it previews the full outgoing payload for approval before any public mutation.

What happens when CI fails or reviews are still pending?

The Skill verifies that check results belong to the current head and proves each failure unrelated or unavailable before concluding. If review or CI remains pending, it reports the head SHA, pending items, and exact resumption point instead of declaring success.