CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cf-sdlc-doc-design

Invoke when the user asks to author, write, revise, generate, or produce a DESIGN or system/technical design — e.g. "generate DESIGN", "design the system", "define components / interfaces / architecture / boundaries". Thin preset binding the DESIGN artifact KIND, delegating authoring and review to the core cf-write-docs engine.

56

Quality

70%

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 ./.claude/skills/cf-sdlc-doc-design/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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 terse, machine-generated bootstrap that clearly delegates all substantive work to external protocol files. It is commendably brief and its stop-on-failure rule is explicit, but it offers no executable guidance, relies on an unresolved path placeholder, and provides no headers or overview to orient the reader.

Suggestions

Explain how {cf-studio-path} is resolved (e.g. an env var or a concrete default path) so the LOAD directives are actionable without guessing.

Consolidate the duplicate ALWAYS rules between the CF_WORKFLOW_ACTIVE block and GeneratedFollowProtocol into one rule set to remove restatement.

Add a short header or 1-2 line summary at the top stating what this skill does and where the real workflow lives, so the body is navigable rather than a flat protocol dump.

DimensionReasoningScore

Conciseness

The body is a lean ~30-line protocol with no padding and no explanation of concepts Claude already knows; every line is a directive. It falls short of the level-5 'every token earns its place' anchor because of repetition (three near-identical 'ALWAYS' RULES lines across the two units, and the meta-control block largely restates what the pdsl units already enforce).

4 / 5

Actionability

Guidance is semi-concrete: it names exact files to LOAD (required-bootstrap.md, doc-design.md), but the paths hinge on an unresolved placeholder ({cf-studio-path}) and the pdsl block is a pseudo-DSL rather than executable instructions, leaving key details (how to resolve the path, what RequiredBootstrap does) missing. This matches 'Some concrete guidance but incomplete' rather than level 4's mostly-executable anchor.

3 / 5

Workflow Clarity

A coherent dispatch sequence is present (load bootstrap → run RequiredBootstrap → load doc-design.md as controlling protocol, traverse in order, evaluate gates), and one explicit checkpoint exists ('ALWAYS stop if any required fragment or rule cannot be followed'). However, the actual design-authoring steps are entirely deferred to external files not shown here, so checkpoints beyond the single stop rule are implicit, matching 'Steps listed but validation gaps'.

3 / 5

Progressive Disclosure

References to external material are one level deep and clearly signaled via LOAD directives, and the body is short. But there are no section headers or navigational structure — it is a flat rule/protocol dump — and the two referenced paths use a placeholder variable and are not part of any bundled files (no references/ or scripts/ exist), so navigation depends on files that cannot be verified. This lands between 'minimal structure' and 'good structure', closer to the level-3 anchor.

3 / 5

Total

13

/

20

Passed

Description

70%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 well-structured trigger-first description with explicit example phrases and a clear scope. Its main weaknesses are jargon-heavy phrasing of what the skill does ('thin preset binding the DESIGN artifact KIND', 'cf-write-docs engine') and near-synonymous verbs standing in for a distinct capability list.

Suggestions

State the what in user-visible terms before the mechanism, e.g. 'Writes and revises system/technical DESIGN documents (components, interfaces, architecture, boundaries); implemented as a thin preset over the cf-write-docs engine.'

Add a few missing natural trigger synonyms such as 'draft a design', 'write the tech design / spec', or 'produce an HLD/LLD' to push trigger coverage toward comprehensive.

Trim the internal jargon ('Thin preset binding the DESIGN artifact KIND') to one plain-language clause so a reader can instantly tell what output the skill produces.

DimensionReasoningScore

Specificity

The description names the domain (DESIGN/system/technical design) and concrete actions ("author, write, revise, generate, or produce a DESIGN"), but the verbs are near-synonyms for one capability rather than a list of distinct actions, and the actual work is opaque behind "delegating authoring and review to the core cf-write-docs engine". This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive' more than level 4, which expects several distinct specific actions with only minor coverage gaps.

3 / 5

Completeness

Both 'when' ("Invoke when the user asks to author, write, revise..." with concrete example phrases) and 'what' (a thin preset binding the DESIGN artifact kind, delegating authoring/review to cf-write-docs) are present. Not a 5 because the 'what' leans on internal jargon ('Thin preset binding the DESIGN artifact KIND') that describes mechanism more than the user-visible outcome, leaving the anchor-5 requirement of a plainly explicit what-and-when not fully met.

4 / 5

Trigger Term Quality

Strong natural trigger coverage: "author, write, revise, generate, or produce a DESIGN", plus example phrases users would actually say ("generate DESIGN", "design the system", "define components / interfaces / architecture / boundaries"). A few common variations are missing (e.g. 'draft', 'spec', 'technical design doc', 'HLD/LLD'), so it sits just below the comprehensive level-5 anchor.

4 / 5

Distinctiveness Conflict Risk

The DESIGN keyword and its quoted trigger phrases carve out a clear niche that distinguishes it from sibling doc-kind skills, but phrases like "design the system" and "define components / interfaces / architecture / boundaries" carry some overlap risk with generic architecture/design skills, matching 'Mostly distinct; minor overlap risk'.

4 / 5

Total

15

/

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

allowed_tools_field

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

Warning

Total

15

/

16

Passed

Repository

constructorfabric/gears-frontx

Path

.claude/skills/cf-sdlc-doc-design/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.