CtrlK
BlogDocsLog inGet started
Tessl Logo

improve-codebase-architecture

扫描代码库中的深化机会,生成可视化 HTML 报告,然后围绕你选中的候选项继续追问。

52

Quality

57%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/engineering/improve-codebase-architecture/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%

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

The body is a well-organized, token-efficient process that assumes Claude's intelligence and gives concrete tooling commands, but it delegates its core HTML artifact to a missing bundle file and lacks explicit output-verification checkpoints.

Suggestions

Provide the HTML report scaffold inline in the body, or ensure HTML-REPORT.md actually ships in ./references/ so the signaled reference resolves.

Add an explicit verification checkpoint after writing the HTML report (e.g. confirm the file exists at the absolute path and opened successfully before asking the user to pick a candidate).

Add a verification/confirm step before the consequential side-effects in the grilling loop (CONTEXT.md updates and ADR proposals) so changes are validated rather than fire-and-forget.

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence — it borrows vocabulary (module, interface, depth, seam, deletion test) from the referenced /codebase-design skill without re-explaining basics, and every section delivers actionable guidance rather than padded concept exposition.

3 / 3

Actionability

Process guidance is concrete (Agent tool with subagent_type=Explore, git log --oneline, $TMPDIR//tmp fallback, xdg-open/open/start, <tmpdir>/architecture-review-<timestamp>.html, Tailwind/Mermaid via CDN), but the central deliverable — the HTML scaffold — is delegated to [HTML-REPORT.md](HTML-REPORT.md) which is not present in the bundle, leaving key executable detail missing.

2 / 3

Workflow Clarity

A clear 3-step sequence (Explore → Present → Grilling) with decision gates (deletion test, '现在不要提出 interfaces', user selection gate, ADR-conflict gating) exists, but there is no explicit verification checkpoint for the generated report or for the consequential side-effects (CONTEXT.md / ADR writes).

2 / 3

Progressive Disclosure

Structure is clean and the HTML-REPORT.md reference is well-signaled and one level deep, but scored against the actual bundle: no references/scripts/assets directories exist and HTML-REPORT.md is absent, so the signaled navigation leads to a missing file.

2 / 3

Total

9

/

12

Passed

Description

50%

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

The description states concrete capabilities but relies on skill-internal jargon ("深化机会") and never gives an explicit use-when trigger. It is adequately specific but not yet distinctive or trigger-complete.

Suggestions

Append an explicit 'Use when...' clause, e.g. 'Use when reviewing codebase architecture for refactoring/deepening opportunities or when the user asks for an architecture review.'

Replace or augment the jargon term "深化机会" with natural trigger terms users would actually say (architecture review, refactoring opportunities, deep vs shallow modules).

Name "architecture" explicitly in the description so the skill is distinguishable from general code-improvement skills.

DimensionReasoningScore

Specificity

Names the domain and several actions — "扫描代码库中的深化机会,生成可视化 HTML 报告...继续追问" — but "深化机会" (deepening opportunities) is abstract jargon rather than the crisp concrete-action list (e.g. "extract text, fill forms, merge documents") that defines a 3.

2 / 3

Completeness

It clearly states what the skill does (scan, generate report, grill) but has no "Use when..." clause or equivalent explicit trigger guidance, which per the rubric caps completeness at 2.

2 / 3

Trigger Term Quality

Relevant keywords are present (codebase, HTML 报告, 候选项, 追问) but the headline trigger "深化机会" is skill-internal jargon a user would rarely say; common natural variations like "architecture review" or "refactoring opportunities" are missing.

2 / 3

Distinctiveness Conflict Risk

The HTML-report-plus-grilling combination is a recognizable niche, but the description omits "architecture" so "扫描代码库中的深化机会" could overlap with generic code-improvement skills and is not maximally distinct.

2 / 3

Total

8

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
vinvcn/mattpocock-skills-zh-CN
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.