CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

improve-codebase-architecture

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/improve-codebase-architecture/SKILL.md

The canonical home for this skill is improve-codebase-architecture in mattpocock/skills

SKILL.md
Quality
Evals
Security

Quality

Content

71%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, dense instruction body with a clear three-phase workflow, concrete commands, and good token discipline. Its one material defect is that the sole bundle reference, HTML-REPORT.md, is missing from the skill directory, so the report-generation detail the body depends on is unreachable.

Suggestions

Add the missing HTML-REPORT.md file to the skill bundle (or remove the reference and inline the essential scaffold), since the "Present candidates as an HTML report" phase depends on it.

Specify which sub-agent type/prompt to spawn for the exploration phase so the scan step is fully executable.

Add a fallback instruction for when the temp HTML file cannot be written or opened (e.g., print the path and continue) to close the workflow's one validation gap.

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence, delegating design vocabulary to the codebase-design skill instead of re-explaining it, with concrete commands kept short. Minor editorial padding ("Be visual.", rationale asides like the YAGNI explanation) keeps it short of anchor 5.

4 / 5

Actionability

Gives concrete, executable guidance: `git log --oneline`, the `$TMPDIR` fallback chain, the exact output path pattern `<tmpdir>/architecture-review-<timestamp>.html`, per-OS open commands, enumerated card fields, and the exact user prompt to ask. A few directions remain abstract ("spawn a sub-agent" without a type or prompt, "widen the net"), which holds it below anchor 5.

4 / 5

Workflow Clarity

Three clearly sequenced phases (Explore, Present report, Grilling loop) with an explicit decision checkpoint ("Do NOT propose interfaces yet... ask the user") and ADR-conflict handling; no destructive or batch operations, so the validation cap does not apply. Missing a feedback loop for the HTML write/open step keeps it below anchor 5.

4 / 5

Progressive Disclosure

The body is well-sectioned and correctly pushes the HTML scaffold detail to a clearly signaled one-level-deep reference ([HTML-REPORT.md](HTML-REPORT.md)), but that file does not exist in the bundle. Scored against the actual bundle structure, the navigation path to the promised detail is broken, which is more than a minor organization gap.

3 / 5

Total

15

/

20

Passed

Description

58%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.

The description states concrete, sequenced actions but reads as internal vocabulary ("deepening opportunities", "grill through") rather than user-facing trigger language. Its main deficiency is the complete absence of a "Use when..." clause, so a model or user matching this skill has no explicit trigger conditions to go on.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to improve codebase architecture, reduce coupling, or find refactoring opportunities."

Replace or gloss internal jargon with natural terms users would say: "refactors that turn shallow modules into deep ones" instead of "deepening opportunities", and "walk through the chosen option" instead of "grill through".

Include common synonyms such as "architecture", "refactor", and "code structure" to raise trigger-term coverage.

DimensionReasoningScore

Specificity

Lists three concrete actions ("Scan a codebase", "present them as a visual HTML report", "grill through whichever one you pick") with minor gaps. "Deepening opportunities" and "grill through" are internal jargon that keeps coverage short of the anchor-5 'comprehensive' level.

4 / 5

Completeness

The "what" is clear (scan, present an HTML report, then walk through the pick), but there is no "Use when..." clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Relevant keywords like "codebase" and "HTML report" are present, but the natural phrases a user would actually say ("architecture", "refactor", "improve the code") are missing. Keyword coverage exists but lacks common variations and synonyms.

3 / 5

Distinctiveness Conflict Risk

The combination of "deepening opportunities", "visual HTML report", and "grill" forms a mostly distinct niche. Minor overlap risk remains with closely related code-review and refactoring skills because "Scan a codebase" is broad.

4 / 5

Total

14

/

20

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

tonykay/transcriber

Path

.agents/skills/improve-codebase-architecture/SKILL.md

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.