skill-builder

Build, audit, and improve Claude Code skills through guided interview and generation. Use when user says "create a skill", "build a skill", "make a new skill", "write a skill", "improve this skill", "audit this skill", "review my skill", "skill from scratch", "turn this into a skill", or /skill-builder. Produces guide-compliant skill directories with SKILL.md, references, and scripts.

Quality

92%

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

85%Weight 40%Scale 1-3

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

The body is well-structured and actionable with strong workflow sequencing and validation feedback loops, and it uses one-level-deep references that all resolve to real files. The main weakness is redundancy between the workflow phases and the later Generation Rules / Reference Files Index sections.

Suggestions

Consolidate the standalone "Generation Rules" section into the Create Mode phases (or convert it to a pure pointer table) to remove the duplication of frontmatter, body, and pattern rules.

Replace the prose "Reference Files Index" table with inline links at first mention so each reference is signaled once rather than described twice.

Trim repeated file-rule restatements (e.g. "No XML tags anywhere" appears in both Phase 3 and File Rules) to a single canonical location.

Dimension	Reasoning	Score
Conciseness	The body is imperative and largely lean, but the later "Generation Rules" and "Reference Files Index" sections restate content already covered in the Create/Edit Mode phases (frontmatter rules, body rules, pattern list, file paths), so it could be tightened — matching the score-2 anchor of mostly-efficient-but-with-redundancy rather than the every-token-earns-its-place score 3.	2 / 3
Actionability	Guidance is concrete and executable: "mkdir -p {scope}/{name}/references", specific frontmatter constraints, "If X fails, do Y" error clauses, and a 3-should/2-should-not test-prompt recipe; per the scoring notes, an instruction-only skill with actionable guidance scores 3 even without code.	3 / 3
Workflow Clarity	Both modes are numbered into phases with explicit validation checkpoints ("Run through every item in references/checklist.md — report pass/fail") and feedback loops (re-run the checklist after changes, confirm previously failing items pass), matching the score-3 anchor for clear sequences with validation and error-recovery loops.	3 / 3
Progressive Disclosure	A clear overview (two modes) precedes detailed phases, a quick-reference section, and a Reference Files Index table; all four referenced files (checklist.md, description-guide.md, frontmatter-spec.md, patterns.md) exist under references/ and are one level deep with signaled navigation, matching the score-3 anchor.	3 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

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: concrete actions, explicit outputs, a rich set of natural trigger phrases, and a clear what+when structure. It uses the bare imperative form seen in the rubric's good examples rather than first/second-person pronouns, so no voice penalty applies.

Dimension	Reasoning	Score
Specificity	"Build, audit, and improve Claude Code skills through guided interview and generation" plus "Produces guide-compliant skill directories with SKILL.md, references, and scripts" lists multiple concrete actions and a concrete output, matching the score-3 anchor; it is not vague like the score-2 example which omits output specifics.	3 / 3
Completeness	It states what the skill does and explicitly answers when to use it via "Use when user says…", satisfying the score-3 anchor for both what AND when with explicit triggers; it is not missing the when clause (which would cap it at 2).	3 / 3
Trigger Term Quality	Nine quoted natural trigger phrases ("create a skill", "build a skill", "audit this skill", "review my skill", "skill from scratch", "turn this into a skill") plus "or /skill-builder" give strong coverage of terms a user would actually say, matching the score-3 anchor rather than the partial coverage at score 2.	3 / 3
Distinctiveness Conflict Risk	The niche (building/auditing Claude Code skills) is clear with distinct, specific triggers unlikely to fire for unrelated skills, matching the score-3 anchor; it is far more specific than the generic score-2 example "Works with document files".	3 / 3
	Total	12 / 12 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: MrToxy/claude-skills
Path: skill-builder/SKILL.md
Commit: 7868606

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.