skill-architect

Architect and audit portable agent skills for retrieval, predictability, progressive disclosure, evals, and repository gates. Use when asked to "create a skill", "author a skill", "improve a skill", "review a skill", "refactor a skill", "compare skill rubrics", "distill a skill from sessions", "reconcile skill plans", or "teach skill authoring"; when paths include `skills/<name>/SKILL.md`, `references/*.md`, helper scripts, evals, lockfiles, security manifests, or vendor adapters; or when mentions include skill-creator, Skill Development, skills.sh, trigger descriptions, completion criteria, leading words, or model/user invocation.

Quality

88%

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%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 well-structured, actionable meta-skill with clear sequenced workflows, validation feedback loops, and clean one-level-deep progressive disclosure. Its main weakness is conciseness — the core workflow is described three times across separate sections.

Suggestions

Collapse the redundant workflow descriptions: "Operating mode", "Skill architecture workflow", and the "Review before patching" predictability pass restate the same ordered steps — keep one canonical sequence and cross-reference it.

Tighten the prose in "Mode rules" and "Deterministic checks vs judgment" into tighter table rows or bullets; several sentences repeat concepts already captured in the command-mode and layer tables.

Move the lengthy self-eval scenario catalog note (the "Authors/reviewers evaluating Skill Architect itself only" paragraph) entirely into references or assets/evals to keep SKILL.md focused on routing.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient domain methodology rather than concepts Claude already knows, but the workflow is restated across "Operating mode", "Skill architecture workflow", and "Review before patching" (the 7-step predictability pass), and several prose sections could be tightened.	3 / 5
Actionability	Concrete, executable guidance throughout — specific commands (`npm run check:skill-standard`, `node bin/hash-check.mjs`), named helper scripts with a use-case table, and a checkable audit-packet table; minor gaps are abstract planning prose where concrete commands would be inappropriate.	4 / 5
Workflow Clarity	Multi-step processes are clearly sequenced (numbered 1–6 workflow, 7-step operating order) with explicit validation checkpoints, fixture-run feedback loops ("an audit is incomplete until each embedded-code finding... includes a fixture command plus observed output"), and STOP/drift conditions.	5 / 5
Progressive Disclosure	SKILL.md is a routing overview with a "Reference files" load-when table; detail is split one level deep into the five real `references/*.md` files, four `assets/scripts/` helpers, and `assets/evals/scenarios.json`, all linked and confirmed present.	5 / 5
	Total	17 / 20 Passed

Description

100%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, third-person description with a concrete capability clause and an explicit, multi-branch 'Use when' trigger covering natural phrases, paths, and mentions. It fully answers both what and when with comprehensive, low-conflict trigger terms.

Dimension	Reasoning	Score
Specificity	Names the domain and lists multiple concrete actions — "Architect and audit portable agent skills for retrieval, predictability, progressive disclosure, evals, and repository gates" — plus the nine command modes (create, author, improve, review, refactor, compare, distill, reconcile, teach), giving comprehensive coverage.	5 / 5
Completeness	Explicitly answers both what ("Architect and audit portable agent skills...") and when ("Use when asked to...") with concrete trigger phrases, satisfying the top completeness anchor.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage with synonyms a user would actually say ("create a skill", "author a skill", "review a skill", "distill a skill from sessions") plus path tokens (`skills/<name>/SKILL.md`, `references/*.md`) and mention terms (skill-creator, skills.sh).	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (skill authoring/architecture) with distinct, specific triggers and path/mention signals, giving minimal conflict risk with unrelated skills.	5 / 5
	Total	20 / 20 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
relative_links	Relative link issues: 6 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 3 missing, 11 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: t4sh/skills4sh
Path: skills/skill-architect/SKILL.md
Commit: d56af8b

Reviewed: 10 hours 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.