aif-commit

Create conventional commit messages by analyzing staged changes. Generates semantic commit messages following the Conventional Commits specification. Use when user says "commit", "save changes", or "create commit".

Quality

83%

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

Quality

Content

77%

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

A highly actionable, well-sequenced skill with strong validation feedback loops, but it suffers from verbosity and structural duplication between the Workflow and Important sections, and keeps complex plan-resolution logic inline rather than offloading it to reference files.

Suggestions

Consolidate the duplicated staging/grouping logic: the 'Important' split-into-separate-commits steps repeat Workflow step 3's hunk-level staging guidance — merge them into one authoritative process to remove the second ~20-line block.

Tighten the skill-context section: drop the meta explanation of why overrides win ('same principle as nested CLAUDE.md files') and state the override rule in one or two lines.

Extract the detailed active-plan resolution algorithm (branch-stem lookup, sequential glob patterns, ultra-marker validation) into a reference file and keep SKILL.md as an overview pointing to it, improving both conciseness and progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly actionable and domain-specific, but padded in places: the skill-context section explains principles Claude already knows ('same principle as nested CLAUDE.md files') and the split/unrelated-work logic in 'Important' duplicates the staging/grouping guidance from Workflow step 3.	2 / 3
Actionability	Highly concrete with real executable git commands (git diff --cached, git add -p, git apply --cached, git push -u origin, etc.), concrete format templates, multiple worked examples, and copy-paste-ready AskUserQuestion blocks.	3 / 3
Workflow Clarity	Multi-step sequences with explicit validation checkpoints and error-recovery feedback loops ('stop before changing staging and ask the user', context gates, confirm-before-commit) for the destructive staging/commit operations, so no missing-validation cap applies.	3 / 3
Progressive Disclosure	Organized into clear sections (Workflow, Format, Examples, Behavior, Important) but is a monolithic single file with no skill-bundle references; the detailed plan-resolution and hunk-staging algorithms are inline walls of edge-case detail that could be split out.	2 / 3
	Total	10 / 12 Passed

Description

90%

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

A strong description that clearly states the capability, follows the Conventional Commits spec, and provides explicit natural trigger terms. The only weakness is that the two action clauses are repetitive rather than enumerating distinct capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and concrete actions ('Create conventional commit messages by analyzing staged changes', 'Generates semantic commit messages following the Conventional Commits specification'), but these are two phrasings of the same capability rather than a list of multiple distinct actions like the score-3 anchor.	2 / 3
Completeness	Both 'what' (create/analyze conventional commit messages from staged changes) and 'when' (explicit 'Use when...' clause with trigger terms) are clearly present, matching the score-3 anchor; voice is third person, no penalty.	3 / 3
Trigger Term Quality	Explicit natural trigger terms users would actually say are given: 'Use when user says "commit", "save changes", or "create commit"' — these are exactly the phrasings a user reaches for when needing a commit.	3 / 3
Distinctiveness Conflict Risk	Conventional commit messages from staged changes is a clear niche with distinct git-oriented triggers; 'save changes' is slightly generic but the overall framing is specific and unlikely to conflict with unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: lee-to/ai-factory
Path: skills/aif-commit/SKILL.md
Commit: 4b943ec

Reviewed: 4 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.