git

Apply Epicenter Git conventions for commits, branches, and history. Use when staging files, writing or splitting commits, creating branches, or inspecting history, not for pull request text or merge guidance.

Quality

86%

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

81%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.

A highly actionable, well-bounded conventions skill: exact format rules, concrete examples, an unusual org-specific versioning scheme, and explicit exclusions of watermarks and PR-scope work. Its main weakness is token efficiency — roughly a third of the body re-teaches standard Conventional Commits that Claude already knows, which also slightly bloats what should be a lean inline skill.

Suggestions

Cut the generic Conventional Commits explanation (the full type list, 'When to Use Scope' / 'When NOT to Use Scope', and 'Other Best Practices' sections) down to a one-line reminder plus only the deltas from the standard spec (e.g., 'Standard Conventional Commits, except:' followed by the 8.Y.Z scheme and `!` semantics).

Merge the scope guidance into a short list of 2-3 rules with one example scope, removing the 'Your current practice is good' coaching sentence and the three separate when/when-not subsections.

Consolidate the watermark prohibition into a single compact rule ("Never include tool/AI attribution or watermarks, including Co-Authored-By trailers") instead of a six-bullet list that enumerates each tool variant.

Dimension	Reasoning	Score
Conciseness	The body spends roughly 40 lines on standard Conventional Commits material Claude already knows ("feat: New features", "Use imperative mood ('add' not 'added' or 'adds')", "When to Use Scope" / "When NOT to Use Scope", "Write commits for future developers"). The genuinely additive content — the unified "8.Y.Z" versioning scheme, the "!" patch/minor semantics, and the watermark prohibitions — is diluted by this re-explanation, matching the "mostly efficient but includes some unnecessary explanation" anchor; it is not the severely padded 2 anchor because nothing is fluffy or repetitive.	3 / 5
Actionability	Concrete format template, six concrete commit examples ("feat(transcription): add model selection for OpenAI providers", "fix!: change default transcription API endpoint"), a good/bad body pair explaining the why, and fully specific rules (lowercase after colon, no period, under 50-72 characters, "Include BREAKING CHANGE: in the commit footer"). Specific examples cover the common cases; nothing is pseudocode or vague.	5 / 5
Workflow Clarity	This is a single-task conventions skill with no multi-step or destructive process, so the simple-skill exception applies: the single action (write a commit following these rules, atomically, splitting via the referenced standalone-commits skill) is unambiguous. The "consider splitting the commit" and explicit PR-boundary delegation remove any ambiguity; the destructive/batch validation cap does not apply.	5 / 5
Progressive Disclosure	No bundle files exist, and the body explicitly declares its structure choice ("This skill keeps its rules inline") with one-level-deep, clearly-signaled cross-skill links to pull-request and standalone-commits. This is the "good structure; most content is appropriately placed" 4 anchor rather than 5, because ~40 lines of generic Conventional Commits reference inlined in SKILL.md could be trimmed since Claude already knows the standard spec.	4 / 5
	Total	17 / 20 Passed

Description

87%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.

A strong description: third-person, concise, explicit what-and-when with concrete trigger phrases, plus a negative scope clause that prevents mis-triggering on PR/merge work. The only minor gaps are a few missing natural synonyms (e.g., "commit message", "amend").

Dimension	Reasoning	Score
Specificity	"Apply Epicenter Git conventions for commits, branches, and history" names the domain and several concrete targets (commits, branches, history). It lists several specific areas rather than performing enumerated actions, so it sits at the 4 anchor and clearly below the comprehensive multi-action 5 anchor.	4 / 5
Completeness	It explicitly answers both questions: the what is "Apply Epicenter Git conventions for commits, branches, and history" and the when is "Use when staging files, writing or splitting commits, creating branches, or inspecting history" with concrete trigger phrases. This matches the anchor-5 example structure exactly; the added exclusion clause only strengthens it.	5 / 5
Trigger Term Quality	"staging files, writing or splitting commits, creating branches, or inspecting history" are natural phrases a user would say, giving good keyword coverage. A few common variations are missing (e.g., "commit message", "amend", "git log"), which keeps it below the comprehensive 5 anchor.	4 / 5
Distinctiveness Conflict Risk	A clear niche (repo-specific Git commit/branch/history conventions) with an explicit negative boundary — "not for pull request text or merge guidance" — that separates it from the sibling pull-request skill. Triggers are distinct with minimal conflict risk.	5 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 suspicious	Warning

	Total	15 / 16 Passed

Repository: EpicenterHQ/epicenter
Path: .agents/skills/git/SKILL.md
Commit: f9441c8

Reviewed: 3 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.