CtrlK
BlogDocsLog inGet started
Tessl Logo

kiro-spec-design

Generate comprehensive technical design translating requirements (WHAT) into architecture (HOW) with discovery process. Use when creating architecture from requirements.

60

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 ./tools/cc-sdd/templates/agents/claude-code-skills/skills/kiro-spec-design/SKILL.md
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.

The skill body is a well-structured, action-oriented workflow: six sequenced steps, explicit validation gates with bounded repair loops, concrete error handling, and correct delegation of detail to rule files. Its weaknesses are minor — some repeated phrasing, and references to bundle files that are not present in the reviewed directory.

DimensionReasoningScore

Conciseness

The body is dense imperative bullets with no explanation of concepts Claude already knows, but there is minor repetition — "Read and apply rules/... from this skill's directory" appears five times, research.md persistence is described in both Step 2 and Step 6, and lines like "Critical: This phase ensures design is based on complete, accurate information" are padding. Anchor 4 (efficient, minor trimming possible), not 5.

4 / 5

Actionability

Concrete file paths ({{KIRO_DIR}}/specs/{feature}/design.md), exact spec.json metadata fields to set, copy-paste user messages, and runnable slash commands (/kiro-spec-design {feature} -y) make the guidance mostly executable. It falls short of 5 because the actual design-authoring guidance is delegated to templates and rule files not present in the bundle, and some directives like "Analyze the codebase to determine which files need to be created vs. modified" remain high-level.

4 / 5

Workflow Clarity

Six clearly sequenced steps with explicit validation checkpoints and feedback loops: approval validation with a stop condition in Step 1, a bounded review-gate loop in Step 5 ("at most 2 repair passes", stop and return to requirements on a real gap), and write-only-after-gate-passes in Step 6. Safety & Fallback enumerates error scenarios with exact stop conditions, user messages, and suggested actions. This matches the anchor-5 example's structure of validate-fix-retry with explicit gating.

5 / 5

Progressive Disclosure

The body delegates detailed guidance to clearly signaled, one-level-deep rule files (rules/design-principles.md, rules/design-discovery-full.md, rules/design-discovery-light.md, rules/design-synthesis.md, rules/design-review-gate.md) at their point of use — a good split. However, none of these files (nor any references/, scripts/, or assets/ bundle) exists in the skill directory as reviewed, so the references cannot be verified and navigation depends on unshipped files. Anchor 4, not 5.

4 / 5

Total

17

/

20

Passed

Description

62%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 answers both what and when in third person with a clear niche, but it stays at a high level: the capability is summarized rather than enumerated, and the trigger vocabulary is narrow. It reads as a competent but improvable description.

Suggestions

Enumerate 2-3 concrete actions in the 'what' clause, e.g. "Generate a technical design document (design.md) covering architecture, components, file structure plan, and testing strategy from approved requirements".

Broaden trigger terms to cover natural user phrasings: "technical design", "design doc", "system architecture", or "turn requirements into a design".

Make the 'when' clause more specific about the entry condition, e.g. "Use when requirements are approved and the user wants to create the technical design or architecture for a feature".

DimensionReasoningScore

Specificity

"Generate comprehensive technical design translating requirements (WHAT) into architecture (HOW) with discovery process" names the domain and the core action, but "comprehensive" and "with discovery process" are broad strokes rather than enumerated concrete actions (no mention of producing a design document, components, file structure plan, or testing strategy).

3 / 5

Completeness

Both a clear 'what' ("Generate comprehensive technical design...") and an explicit "Use when creating architecture from requirements" trigger are present, but the 'when' clause is a single narrow phrase and could be more specific about entry conditions.

4 / 5

Trigger Term Quality

"technical design", "requirements", "architecture", and "creating architecture from requirements" are natural trigger terms, but common variations users would actually say — "design doc", "spec", "system design", "tech design" — are missing.

3 / 5

Distinctiveness Conflict Risk

The requirements-to-design pipeline is a clear niche with a concrete trigger, but the generic "architecture"/"design" wording could overlap with general design or architecture-planning skills.

4 / 5

Total

14

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

frontmatter_unknown_keys

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

Warning

Total

13

/

16

Passed

Repository
gotalab/cc-sdd
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.