CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

tabletopkit

Builds multiplayer spatial board games with TabletopKit on visionOS. Use for tables, seats, equipment, turns, TabletopAction state changes, interactions, snapping/tossing/physics, RealityKit rendering, deterministic game state, or Group Activities synchronization.

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

78%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-architected routing skill: dense, symbol-precise guidance with no filler, a sequenced workflow with verification, a review checklist, and a correctly structured one-level reference bundle. The main improvements are precision — the "availability matrix below" pointer misdirects since the matrix lives in a reference file — and closing the loop between the workflow's verification step and the Common Mistakes remedies.

Suggestions

Fix the dangling pointer "The availability matrix below owns version details" — no matrix appears in the body; point precisely to the Availability notes in references/tabletopkit-patterns.md so readers can actually locate version requirements.

Close the feedback loop in Workflow step 5: state what to do when verification fails (e.g., "on rollback or snapshot mismatch, reconcile via TabletopGame.Observer and re-run the checklist") instead of leaving recovery implicit in Common Mistakes.

Inline one small copy-paste snippet for the most common setup path (seat claiming plus setup.register(action:)) so the body's guidance is executable without opening a reference, and drop the Contents TOC, which duplicates the section headers and Route by Task at this length.

DimensionReasoningScore

Conciseness

The ~80-line body assumes Claude's competence — no explanation of what visionOS, RealityKit, or SharePlay are — and every section carries information. Minor trimmable content exists: the Contents TOC is redundant with headers and Route by Task for this length, and Core Decisions items ("Treat SharePlay transport, reconnection, and late joining as lifecycle states") partially restate Workflow step 5 and checklist items. Efficient with minor trim opportunities fits anchor 4; not 5 because a few tokens do not earn their place.

4 / 5

Actionability

Highly concrete symbol-level guidance throughout: "claimAnySeat() or claimSeat(_:)", "setup.register(action:)", "actionWasRolledBack(_:snapshot:)", "outcome.tossableRepresentation.face(for: outcome.restingOrientation)", ".tabletopGame(_:parent:automaticUpdate:)" on RealityView, plus hard environmental facts ("Group Activities do not work in Simulator", "multiplayer requires 2+ Apple Vision Pro units on FaceTime"). Not 5: no executable code snippet lives in the body, and "The availability matrix below owns version details" points to a matrix that is not below — it resides in references/tabletopkit-patterns.md's Availability notes — leaving one routing instruction imprecise.

4 / 5

Workflow Clarity

The five-step workflow is clearly sequenced with an opening checkpoint ("Confirm visionOS availability, TabletopKit capability, and whether the experience needs local, shared, or SharePlay-backed play") and an explicit verification step ("Verify reset, reconnect, late-join, undo, and multi-participant behavior"), reinforced by a 12-item review checklist. Falls between anchors 4 and 5: error-recovery knowledge exists (Common Mistakes covers rollback, discarded actions, invalid parentID) but is not looped back into the workflow as "if X fails, do Y" steps, so the anchor-5 feedback-loop criterion is only implicitly met.

4 / 5

Progressive Disclosure

A genuine entrypoint: the body stays an overview while 1,277 lines of detail are split across three real, one-level-deep reference files, each signaled twice with descriptions (Route by Task and References), and cross-links between reference files are sibling-level, not nested. Verified against the actual bundle — all three referenced paths (references/core-implementation.md, references/tabletopkit-patterns.md, references/tabletop-networking-and-state.md) exist and carry the promised content. Clear overview, well-signaled one-level references, easy navigation matches anchor 5; the misleading "below" pointer to the availability matrix is too small to drop it to anchor 4's "references mostly clear".

5 / 5

Total

17

/

20

Passed

Description

92%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, concrete, and explicit about both capability and trigger conditions, with framework-anchored terms that make false triggering very unlikely. The only gap is synonym coverage — "SharePlay" (the user-facing name for Group Activities), "dice", and "cards" would make the trigger vocabulary fully comprehensive.

DimensionReasoningScore

Specificity

"Builds multiplayer spatial board games with TabletopKit on visionOS" plus a ten-item capability list ("tables, seats, equipment, turns, TabletopAction state changes, interactions, snapping/tossing/physics, RealityKit rendering, deterministic game state, or Group Activities synchronization") gives comprehensive, concrete coverage of the domain. Not 4: no notable capability gaps remain unlisted.

5 / 5

Completeness

Explicit third-person 'what' ("Builds multiplayer spatial board games with TabletopKit on visionOS") paired with an explicit trigger clause ("Use for tables, seats, equipment, turns, ...") with concrete trigger phrases. Matches the anchor-5 example structure exactly; not 4 because the 'when' is explicit and enumerated, not merely present.

5 / 5

Trigger Term Quality

Natural user vocabulary is well covered ("TabletopKit", "visionOS", "tables", "seats", "equipment", "turns", "RealityKit", "Group Activities", "board games", "multiplayer"), but common synonyms are missing — users routinely say "SharePlay" rather than "Group Activities", and core domain terms "dice" and "cards" are absent. Good coverage with a few natural terms missing fits anchor 4 better than anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

Triggers are anchored to a single framework and platform (TabletopKit, TabletopAction, visionOS, RealityKit, Group Activities), forming a clear niche with minimal overlap risk against any other skill. Not 4: the framing is framework-specific, not merely "mostly distinct".

5 / 5

Total

19

/

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

thienqo/ios-skills

Path

skills/tabletopkit/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.