gentle-ai-chained-pr
Trigger: PRs over 400 lines, stacked PRs, review slices. Split oversized changes into chained PRs that protect review focus.
68
83%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description with excellent trigger terms and clear distinctiveness. It effectively communicates both when to use the skill and what it does, though the specificity of concrete actions could be slightly more detailed (e.g., mentioning branch management, dependency tracking between stacked PRs). The 'Trigger:' prefix is an effective alternative to 'Use when' for signaling selection criteria.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (PR splitting) and a key action ('split oversized changes into chained PRs'), but doesn't list multiple concrete actions like creating branches, updating base branches, or managing dependencies between stacked PRs. | 2 / 3 |
Completeness | Explicitly answers both what ('Split oversized changes into chained PRs that protect review focus') and when ('PRs over 400 lines, stacked PRs, review slices') with a clear 'Trigger:' clause serving as the 'Use when' equivalent. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would say: 'PRs over 400 lines', 'stacked PRs', 'review slices', 'oversized changes', 'chained PRs'. These cover common variations of how users describe the need to split large pull requests. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche focused specifically on splitting large PRs into stacked/chained PRs. The specific triggers (400+ lines, stacked PRs, review slices) are unlikely to conflict with general code review or git workflow skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, concise skill that clearly defines when and how to split oversized PRs into chained PRs. Its main strength is the tight decision gate table and hard rules that leave little ambiguity. Its primary weakness is the lack of inline concrete examples (git commands, PR body templates) — the actionable details are deferred to a reference file that wasn't provided for verification.
Suggestions
Add at least one inline example of a Chain Context PR body section (the dependency diagram with 📍 marker) so Claude has a copy-paste ready template without needing to load the reference file.
Include a minimal concrete example of the branch/PR creation commands (e.g., `gh pr create --base main --head feature/step-1`) for at least one chain strategy to boost actionability.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section is lean and purposeful. No unnecessary explanations of what PRs are or why code review matters. The decision table and hard rules are tightly written, and each token contributes actionable information. | 3 / 3 |
Actionability | The skill provides clear decision gates and execution steps, but lacks concrete executable examples—no actual git/gh commands for creating branches, no example PR body template inline, no copy-paste ready chain context block. The concrete details are deferred to a reference file. | 2 / 3 |
Workflow Clarity | The six execution steps are clearly sequenced with validation at step 5 (CI/tests/docs/manual checks, rollback scope, clean diff). The decision gate table provides explicit branching logic, and the tracker PR lifecycle (draft until all children integrated) serves as a feedback loop for the feature branch chain strategy. | 3 / 3 |
Progressive Disclosure | There is a single well-signaled reference to chaining-details.md for strategy diagrams and branch commands, which is good structure. However, the bundle file was not provided, making it impossible to verify the reference exists, and the skill defers most of the actionable content (commands, templates, diagrams) to that single file, leaving the main skill somewhat thin on executable detail. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- sergiodvillegas-art/gentle-ai
- Commit
3bfa934
- Reviewed
Table of Contents
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.