CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

swift-concurrency

Diagnoses and implements data-race-safe Swift concurrency with actor isolation, Sendable, structured tasks, cancellation, and synchronization. Use for compiler diagnostics or async behavior; when the requested deliverable is a PR or diff review, select swift-code-review first.

Quality

89%

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

85%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, domain-expert-level skill body: a sequenced diagnostic workflow with validation and a checklist, a concrete isolation decision table, named safety rules with specific APIs, and a clean one-level-deep reference bundle that all resolves to real files. The main residuals are mild redundancy between Safety Rules and Common Mistakes plus a redundant TOC, and the absence of any inline code examples for the highest-risk patterns.

Suggestions

Drop the Contents table-of-contents (section headers are already scannable in a ~100-line file) and merge the overlapping entries between Safety Rules and Common Mistakes — e.g., the lock-across-await and detached-task items appear in both lists — to save tokens.

Add one short inline snippet for the highest-risk pattern (a checked continuation with exactly-once completion on every terminal path), since the body currently defers all code to references/bridging-interop.md.

Signal references/swiftui-concurrency.md from a body section (e.g., from Scope or Isolation Decisions, where SwiftUI .task routing is mentioned) so it is not discoverable only via the flat References list.

DimensionReasoningScore

Conciseness

The body is lean and imperative with no basic-concept explanations ("Never hold a synchronous lock across `await`" assumes competence), but the Contents TOC duplicates the section headers and Common Mistakes restates several Safety Rules nearly verbatim (lock-across-await, detached tasks, actor revalidation after await). This fits anchor 4's 'minor instances... that could be trimmed' rather than anchor 5's 'every token earns its place'.

4 / 5

Actionability

Concrete guidance throughout: a decision table mapping situations to specific isolation choices, named APIs (`@unchecked Sendable`, `nonisolated(unsafe)`, `Task.detached`, `async let`, `TaskGroup`, `AsyncStream`), and scoped pointers to reference files. However, no inline code examples of the key patterns (exactly-once continuation, stream termination cleanup) exist in the body — they are deferred to references — so it stops short of anchor 5's 'specific examples cover the common cases'.

4 / 5

Workflow Clarity

The Diagnostic Workflow is a clear 5-step sequence with a toolchain preflight, an explicit validation checkpoint ("Rebuild, inspect new diagnostics, and test cancellation/failure behavior affected by the change"), an implicit fix-and-recheck loop, and a closing Review Checklist — matching anchor 5's 'explicit validation steps; feedback loops...; checklists'. This is not a destructive or batch operation, so no cap applies.

5 / 5

Progressive Disclosure

The body is a true overview and all seven referenced files (references/diagnostics.md, approachable-concurrency.md, synchronization-primitives.md, concurrency-patterns.md, bridging-interop.md, async-algorithms.md, swiftui-concurrency.md) exist with substantive content at one level deep, six of them signaled inline with purpose statements (e.g., "Use [Diagnostics](references/diagnostics.md) for compiler-message-specific remedies"). Bulk detail (e.g., the 417-line synchronization guide) is appropriately externalized, matching anchor 5; the only nit is that swiftui-concurrency.md is surfaced only in the flat References list.

5 / 5

Total

18

/

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 in third-person voice that states concrete capabilities, enumerates the five core concern areas of Swift concurrency, provides an explicit 'Use for...' trigger clause, and even includes cross-skill routing. The only improvements are adding a few more natural trigger synonyms and one or two additional action verbs for full comprehensiveness.

DimensionReasoningScore

Specificity

"Diagnoses and implements data-race-safe Swift concurrency with actor isolation, Sendable, structured tasks, cancellation, and synchronization" names several specific capabilities and enumerated technologies, with only minor coverage gaps (e.g., migration, debugging runtime hangs). It is more comprehensive than the 1-2 concrete actions of anchor 3 but does not list the multiple distinct action verbs of anchor 5.

4 / 5

Completeness

It clearly answers 'what' ("Diagnoses and implements data-race-safe Swift concurrency with actor isolation, Sendable...") and explicitly answers 'when' with concrete trigger phrases ("Use for compiler diagnostics or async behavior"), plus routing guidance to swift-code-review. This matches anchor 5; anchor 4 would require the 'when' to be less explicit.

5 / 5

Trigger Term Quality

Natural phrases a user would say are present ("data-race", "actor isolation", "Sendable", "cancellation", "compiler diagnostics", "async behavior"), but common variations like "warning", "error", "async/await", "MainActor", or "Swift 6" are missing, matching anchor 4's 'good keyword coverage; a few natural terms missing' rather than anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

"data-race-safe Swift concurrency", "actor isolation", "Sendable", and "compiler diagnostics" form a clear niche with distinct triggers, and the description proactively disambiguates from swift-code-review ("when the requested deliverable is a PR or diff review, select swift-code-review first"), giving minimal conflict risk per anchor 5.

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

thienqo/ios-skills

Path

skills/swift-concurrency/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.