CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

architect

Sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in. Use for /architect, 'architect this', 'design this', or non-trivial work where jumping to code would lock in the wrong shape.

72

Quality

90%

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

92%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 orchestration skill: lean, directive prose with an explicit phased workflow, genuine feedback loops, and clean one-level-deep reference files verified to exist and carry real content. The only gap is reliance on out-of-bundle sibling skills (how, arena, interrogate, principle skills) for the mechanics of grounding and multi-model synthesis.

Suggestions

Bundle or briefly summarize the contract with the **how** and **arena** skills (expected inputs/outputs of grounding artifacts and the arena task format) so Phase A and Phase B are executable from the skill alone.

Consider linking the other referenced skills (arena, how, why, interrogate) the way design-red-flags.md is linked, so their existence and location are verifiable from the body.

DimensionReasoningScore

Conciseness

The body is lean and directive ('Design it twice.', 'Naming a file isn't grounding.', 'Subtract before adding') with no explanations of concepts Claude already knows; every rationale sentence (interface depth, scrap tells) is load-bearing judgment guidance rather than padding.

5 / 5

Actionability

Directives are concrete where it counts (explicit fallback runner names 'claude-opus-5-5-xhigh' and 'grok-4.7-xhigh-fast', pass references/runner-prompt.md, screen against references/design-red-flags.md, an explicit scrap-tells checklist), but Phase A and Phase B execution depends on out-of-bundle skills (**how**, **arena**) whose mechanics are neither bundled nor summarized, leaving minor gaps.

4 / 5

Workflow Clarity

The sequence is explicit (todolist with one entry per phase, Phases A-E) with real validation checkpoints (red-flag screening before synthesis, Phase C opt-in sign-off, Phase D deviations surfaced) and error-recovery feedback loops (human pushback treated as Phase A evidence triggering re-ground and re-run; scrap criteria looping back to Phase B).

5 / 5

Progressive Disclosure

The body is an overview-length orchestration layer; all three bundle files exist, are substantive, and are clearly signaled as one-level-deep references (linked references/design-red-flags.md, references/rationale-template.md, references/runner-prompt.md), each loaded only at the phase where it is needed.

5 / 5

Total

19

/

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 that clearly states what the skill does and when to use it, with natural trigger phrases in third person. Trigger coverage and capability coverage are good but not exhaustive, and 'design this' has slight overlap with visual-design intents.

Suggestions

Add one or two natural trigger synonyms (e.g., 'plan the structure', 'scaffold the design') to broaden keyword coverage.

Mention the multi-model design-synthesis capability in the 'what' clause to round out capability coverage.

Scope 'design this' to code/module design (e.g., 'design this code') to reduce overlap with visual-design skills.

DimensionReasoningScore

Specificity

Lists several concrete actions ('Sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in') but omits capabilities like multi-model synthesis, red-flag screening, and the scrap loop, so coverage has minor gaps rather than being comprehensive.

4 / 5

Completeness

Explicitly answers both 'what' (sketch types, signatures, module structure, stay in the loop during fill-in) and 'when' ('Use for /architect, "architect this", "design this", or non-trivial work where jumping to code would lock in the wrong shape') with concrete trigger phrases, in third person.

5 / 5

Trigger Term Quality

Includes natural trigger phrases users would say ('/architect', 'architect this', 'design this') plus a use-condition, but misses common synonyms such as 'plan the structure' or 'scaffold the design'.

4 / 5

Distinctiveness Conflict Risk

The architecture-sketch niche is mostly distinct, but the trigger 'design this' carries minor overlap risk with UI/visual-design skills; the slash command and 'architect this' are clearly distinct.

4 / 5

Total

17

/

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

cursor/plugins

Path

pstack/skills/architect/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.