branch-surgery-pr-split

Split oversized or mixed-concern branches into smaller, reviewable PR stacks with safety refs, topology selection, parity audits, and merge sequencing. Use when a PR or branch is too large, difficult to review, mixed across concerns, conflict-prone, or needs to be decomposed without losing net changes.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "branch-surgery-pr-split" with this command: npx skills add liatrio-labs/ai-prompts/liatrio-labs-ai-prompts-branch-surgery-pr-split

Branch Surgery PR Split

Overview

Use this skill to turn a risky monolithic branch into a safe, auditable set of scoped PRs. Treat topology as a decision, not a default recipe, and collaborate with the user before executing branch surgery.

Workflow

  1. Quantify branch shape before proposing solutions.
  2. Always present the top two topology options with explicit tradeoffs.
  3. Collaborate with the user to select one topology.
  4. Produce a decision-complete split plan.
  5. Execute with safety refs and strict scope boundaries.
  6. Run parity and completeness audits.
  7. Remediate drift and generate PR metadata.

Step 1: Quantify Branch Shape

Collect objective signals first:

  • Commit count and net churn versus base.
  • Number of changed files and hotspot files.
  • Mixed-commit prevalence (single commit touching unrelated concerns).
  • Dependency density between concerns.
  • Expected review risk and conflict risk.

If signals are incomplete, gather more evidence before recommending topology.

Step 2: Generate Topology Options

Always provide the top two options with tradeoffs and a recommendation. Use references/topology-patterns.md to select options.

For each option, include:

  • Topology name.
  • Why it fits this branch shape.
  • Review ergonomics, merge complexity, and conflict risk.
  • Failure modes and mitigations.
  • Cost of execution (low/medium/high).

Do not skip tradeoffs.

Step 3: Collaborate on Topology Selection

Use targeted discovery questions from references/facilitated-discovery-questions.md. Confirm constraints before surgery:

  • Must-merge-first dependencies.
  • Parallel review goals.
  • Tolerance for stacked PR depth.
  • Documentation isolation preference.
  • Release timing constraints.

If the user is unsure, recommend one option and explain what would change with the second option.

Step 4: Produce a Decision-Complete Split Plan

Produce:

  • Branch inventory with base branch mapping.
  • Commit disposition map (keep, split, drop-as-superseded).
  • Execution order.
  • Merge order.
  • Validation gates.
  • Risk register and mitigations.

Use templates in references/pr-stack-templates.md.

Step 5: Execute Branch Surgery Safely

Execution rules:

  • Create backup refs first (tag and/or backup branch).
  • Preserve explicit scope contracts per split branch.
  • For mixed commits, use cherry-pick -n and stage only owned files/hunks.
  • Keep shared-doc churn isolated when selected topology calls for it.
  • Avoid hidden carryover changes by checking staged file lists before commit.

Step 6: Audit Before Confidence

Run audit gates from references/audit-gates.md. A split is incomplete until all gates pass or deviations are documented.

Mandatory checks:

  • No missing files compared to original net diff.
  • No unexpected extra files.
  • Per-file parity or intentional documented divergence.
  • Deletion semantics parity.
  • Branch scope boundaries match PR intent.

Step 7: Remediate and Package PRs

If audit finds issues:

  1. Fix scope drift or ownership errors.
  2. Re-run audits.
  3. Remove redundant branches.

Then generate PR metadata and merge sequencing notes using references/pr-stack-templates.md.

Output Standard

Always provide:

  1. Chosen topology and why it won over the runner-up.
  2. Branch plan and merge order.
  3. Audit summary with pass/fail status per gate.
  4. Remediation actions taken.

References

  • references/topology-patterns.md
  • references/facilitated-discovery-questions.md
  • references/audit-gates.md
  • references/pr-stack-templates.md

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

liatrio-brand-guidelines

No summary provided by upstream source.

Repository SourceNeeds Review
General

create-pull-request

No summary provided by upstream source.

Repository SourceNeeds Review
General

git-commit-conventional

No summary provided by upstream source.

Repository SourceNeeds Review
General

create-mermaid-diagrams

No summary provided by upstream source.

Repository SourceNeeds Review