CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

tdd

Red-Green-Refactor for changes with behavior or regression risk. Use when: adding observable behavior, fixing a bug, or changing logic not already covered by a precise executable check. Not for: pure docs/research, deterministic generated-artifact refreshes, or mechanical changes already covered by an existing checker. Output: observed RED (new test or existing failing check) → minimal GREEN → refactor under protection.

Quality

84%

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

SKILL.md
Quality
Evals
Security

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 well-built process skill: lean, table-driven risk routing, explicit feedback loops around the RED/GREEN/REFACTOR cycle, and concrete worked examples. Remaining improvements are minor — trimming the redundant opening policy paragraph and verifying the out-of-bundle capsule reference resolves correctly.

DimensionReasoningScore

Conciseness

The body is dense and assumes competence — a risk-routing table, RED-source rules, an ASCII cycle, and concrete examples ('pnpm check:sop-definitions') with zero padding on concepts Claude already knows. The opening policy paragraph (RED 是证据状态...) restates material that reappears in the table and mistakes sections, leaving minor trimmable redundancy that keeps it below anchor 5.

4 / 5

Actionability

For an instruction-only skill the guidance is highly concrete: per-change-type routing ('现有精确检查已经红 → 直接修到该检查绿'), an exact command example, and worked positive/negative/grey cases covering common situations. It stops short of anchor 5 because most guidance is decision-routing rather than copy-paste-ready commands, and the referenced capsule path cannot be executed from the skill directory.

4 / 5

Workflow Clarity

The RED → GREEN → REFACTOR cycle is explicitly sequenced with validation checkpoints ('复跑同一保护集') and explicit error-recovery feedback loops for each failure mode: RED passing immediately, RED being a syntax/environment error, GREEN breaking other checks, and REFACTOR turning red ('回退这次重构,重新收敛'). This matches anchor 5 exactly.

5 / 5

Progressive Disclosure

A self-contained, well-sectioned ~80-line body with no bundle files — sections are clearly organized and everything is at the right altitude. It falls short of anchor 5 because the single external reference ('../.cat-cafe-shared-refs/bug-diagnosis-capsule.md') points outside the skill directory and cannot be verified against the (empty) bundle, and the body slightly exceeds the size where organization alone earns top marks.

4 / 5

Total

17

/

20

Passed

Description

83%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, with an explicit Use when / Not for boundary structure and a concrete output flow. Main gaps are missing common synonyms ('TDD', 'test-driven') in the description text itself and residual overlap with debugging skills on the bug-fix trigger.

DimensionReasoningScore

Specificity

The description names the domain ('Red-Green-Refactor for changes with behavior or regression risk') and concrete actions ('observed RED (new test or existing failing check) → minimal GREEN → refactor under protection'), distinguishing new-test vs existing-check RED. It falls just short of anchor 5 because the actions are stated at the workflow level without enumerating the concrete operations each phase entails.

4 / 5

Completeness

It explicitly answers both: what ('Output: observed RED ... → minimal GREEN → refactor under protection') and when ('Use when: adding observable behavior, fixing a bug, or changing logic not already covered by a precise executable check'), plus an explicit negative trigger ('Not for: pure docs/research, deterministic generated-artifact refreshes...'). This matches anchor 5 — concrete what and when with explicit trigger phrases — and exceeds anchor 4 because 'when' is fully explicit rather than improvable.

5 / 5

Trigger Term Quality

Natural terms users would say are present: 'fixing a bug', 'regression', 'behavior', 'Red-Green-Refactor'. Common synonyms like 'TDD', 'test-driven', 'write tests first', or 'unit test' are absent from the description itself (they only appear in the triggers frontmatter field, which is not the evaluation target), so it does not reach comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

The 'Not for' clause carves a clear niche against docs/formatting/checker-covered changes, but 'fixing a bug' as a use-when trigger overlaps with debugging-type skills, leaving minor overlap risk with a closely related skill rather than the minimal-conflict profile of anchor 5.

4 / 5

Total

17

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 1 suspicious

Warning

Total

14

/

16

Passed

Repository

zts212653/clowder-ai

Path

cat-cafe-skills/tdd/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.