pr-splitter

Use when breaking a large, complex, messy, or hard-to-review pull request into multiple smaller PRs; planning stacked PRs; extracting independent changes from a branch; splitting mixed refactor and behavior changes; managing drift after review feedback; rebasing follow-up PRs as earlier PRs change; or preserving original branch intent while shipping incrementally.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/pr-splitter/SKILL.md

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a concrete, well-sequenced operational guide with real git commands, safety steps, and templates. Its main weaknesses are redundant restatement of the changeset rule, an internal inconsistency between the changeset instructions and the scratchpad template, and unresolved placeholders/dangling references.

Suggestions

State the 'do not keep the original changeset' rule once (in the Changesets section) and drop its repetition in numbered step 1 and Common failure modes.

Add the changeset line to the scratchpad template itself, since the text instructs 'Add changeset creation to the scratchpad template' but the template omits it.

Resolve the '<base>' placeholder with a concrete way to determine the base branch, and replace the dangling '.mastracode/commands/changeset.md' path with the actual CLI usage inline or a verified link.

Dimension	Reasoning	Score
Conciseness	The body is mostly imperative and dense, but the 'do not keep the original changeset' rule is stated three times (Changesets intro, numbered step 1, and Common failure modes), which is noticeable redundancy rather than a minor trim. It fits the 'mostly efficient but could be tightened' anchor better than the score-4 'minor instances' anchor.	3 / 5
Actionability	Concrete, executable commands appear throughout ('git branch backup/original-large-pr', 'git checkout backup/original-large-pr -- path/to/file', 'git restore -p --source backup/original-large-pr -- path/to/file', 'git range-diff', 'pnpm changeset -s -m "..." pkg-name') plus ready-to-use templates. It falls short of 5 because the '<base>' placeholder is never resolved and '.mastracode/commands/changeset.md' is a dangling path reference.	4 / 5
Workflow Clarity	The 8-step required workflow is clearly sequenced, snapshots the original branch before touching history, and includes an explicit per-PR verification step with a scratchpad checklist. It is not 5 because step 7 (rebasing dependent PRs) never explicitly instructs re-running verification after the rebase, and the changeset instruction to extend the scratchpad template is not reflected in the template shown.	4 / 5
Progressive Disclosure	Sections are well organized for a single-file workflow skill with templates placed inline where they belong, but the repo-specific changeset procedure (~24 lines) is inlined and the one external reference ('.mastracode/commands/changeset.md') is a bare, unverifiable relative path rather than a clearly signaled link. This matches 'good structure; minor organization gaps' rather than the clearly-signaled-references anchor at 5.	4 / 5
	Total	15 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it combines an explicit, multi-trigger 'Use when' clause with concrete capability verbs covering the full PR-splitting lifecycle. Trigger coverage is good though a few community synonyms (e.g. 'stacked diffs') would make it exhaustive.

Dimension	Reasoning	Score
Specificity	The description lists six-plus concrete actions ('breaking a large... pull request into multiple smaller PRs', 'planning stacked PRs', 'extracting independent changes', 'managing drift after review feedback', 'rebasing follow-up PRs', 'preserving original branch intent') with comprehensive coverage of the splitting domain. It exceeds the score-4 anchor because there are no meaningful coverage gaps.	5 / 5
Completeness	An explicit 'Use when...' clause with seven concrete trigger phrases answers the when completely, and the verb phrases ('breaking... into multiple smaller PRs', 'planning stacked PRs', 'managing drift') explicitly state the what. It is not capped at 3 since explicit trigger guidance is present, and the 'when' is more explicit than the score-4 anchor allows.	5 / 5
Trigger Term Quality	Natural phrases users would say are well covered ('hard-to-review pull request', 'stacked PRs', 'review feedback', 'rebasing'), but a few common variants are missing (e.g. 'split up a PR', 'stacked diffs'). Good coverage with minor gaps matches the score-4 anchor.	4 / 5
Distinctiveness Conflict Risk	A clear niche (splitting/stacking PRs) with distinct triggers like 'stacked PRs', 'drift after review feedback', and 'extracting independent changes from a branch' gives minimal overlap with generic git or review skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mastra-ai/mastra
Path: .claude/skills/pr-splitter/SKILL.md
Commit: 3b0d190

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.