CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

atlas

Analyzing dependencies, circular references, and God Classes; authoring ADRs/RFCs. Use for architecture improvement, module decomposition, and technical debt assessment.

Quality

88%

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

82%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 dense, well-structured instruction skill: concrete thresholds, named tooling, explicit decision rules, and a clearly phased workflow with strong reference navigation. The main costs are a duplicated metadata comment block, some trimmable rationale, and inline Multi-Engine detail that duplicates its own reference file.

Suggestions

Trim or collapse the CAPABILITIES_SUMMARY comment block and BIDIRECTIONAL_PARTNERS list, which duplicate the Recipes table and Collaboration section token-for-token.

Move the detailed Multi-Engine Mode mechanics (scoring semantics, degraded-mode rules) fully into reference/tri-engine-architect.md, keeping only activation triggers and the core never-merge rule inline.

DimensionReasoningScore

Conciseness

The body is dense and high-signal (e.g., "<5% healthy, 5-10% significant, >10% critical") with no explanations of concepts Claude already knows. Minor over-explanation exists: the CAPABILITIES_SUMMARY comment block and BIDIRECTIONAL_PARTNERS duplicate the Recipes and Collaboration sections, and some anti-pattern rationale could be trimmed.

4 / 5

Actionability

For an instruction-only skill the guidance is fully actionable: concrete thresholds ("allocate ≥15% of development time"), explicit decision rules ("high-frequency bidirectional → merge candidates; long cycles → extract shared logic"), named tools (ArchUnit, dependency-cruiser, NetArchTest), and defined scoring semantics (CONFIRMED 3/3, CONVERGENT, etc.).

5 / 5

Workflow Clarity

The SURVEY → PLAN → VERIFY → PRESENT table gives a clear sequence with per-phase key rules and a VERIFY validation checkpoint ("YAGNI check, Least Surprise test"). It falls short of 5 because the main workflow lacks explicit error-recovery feedback loops (only multi-engine degraded modes are addressed).

4 / 5

Progressive Disclosure

Navigation is excellent: a Reference Map with a "Read this when" column, per-recipe "Read First" pointers, and one-level-deep references. Minor gaps remain — the inline Multi-Engine Mode section overlaps reference/tri-engine-architect.md, ten dense Core Contract rules could partly live in reference files, and the referenced bundle files are not present in this bundle to verify.

4 / 5

Total

17

/

20

Passed

Description

87%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 that states concrete capabilities in third person and pairs them with an explicit 'Use for' clause listing concrete trigger tasks. Trigger coverage is good but omits a few natural synonyms, and several body capabilities (coupling metrics, C4, boundary evaluation) go unmentioned.

DimensionReasoningScore

Specificity

"Analyzing dependencies, circular references, and God Classes; authoring ADRs/RFCs" plus "module decomposition, and technical debt assessment" lists several concrete actions. It falls short of 5 because body capabilities like coupling metrics, C4 modeling, and boundary evaluation are not covered.

4 / 5

Completeness

Both what ("Analyzing dependencies… authoring ADRs/RFCs") and when ("Use for architecture improvement, module decomposition, and technical debt assessment") are explicitly and concretely answered, mirroring the anchor-5 example structure of concrete trigger phrases.

5 / 5

Trigger Term Quality

Natural terms users would say are present: "dependencies", "circular references", "God Classes", "ADRs", "RFCs", "technical debt". A few natural synonyms are missing (e.g., "coupling", "architecture review", "restructure"), keeping it below comprehensive coverage.

4 / 5

Distinctiveness Conflict Risk

"God Classes", "circular references", and "ADRs/RFCs" carve a clear architecture-analysis niche with distinct triggers and minimal conflict risk; only slight adjacency to refactoring skills via "module decomposition".

5 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

simota/agent-skills

Path

atlas/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.