Swift Code Quality

This documentation was generated with the assistance of AI. Please report any inaccuracies.

Run this Swift package/app’s wired static-analysis and formatting tools — SwiftLint, the Swift toolchain’s bundled swift format lint (or the third-party SwiftFormat tool when that’s what’s configured instead), and Periphery — then report every issue found, grouped by tool and classified by rule id/severity. This is a read-only reporting skill: it runs static analysis and formatting checks and reports results, it never fixes issues or edits any tool’s configuration.

Purpose

iru-swift-code-quality is the Swift-stack equivalent of iru-java-code-quality/iru-typescript-code-quality/ iru-android-code-quality: a fast, focused lint/format/unused-code pass callable on its own or as a baseline capture inside a task group. It verifies each tool is actually wired — a config file present and its binary on PATH — before trusting a "zero issues" result from it, and never installs a missing tool itself, instead offering Homebrew (or Mint/direct-download alternatives) via AskUserQuestion. SwiftLint and swift format lint accept paths directly, so a scoped invocation narrows what actually runs; Periphery has no path selector, so a scoped invocation still runs the whole module-graph scan and only filters the report.

Purpose group

Swift Development Tooling

Plugin

Shipped by the iru-swift plugin of the ai-catalog marketplace (install with /plugin install iru-swift@ai-catalog; see Plugins).

Invocation

/iru-swift-code-quality
/iru-swift-code-quality [scope]

Inputs

Argument Required Description Default

scope

No

A file/directory path or a type/file simple name (e.g. Sources/Foo.swift, Foo). Narrows what SwiftLint and swift format lint actually run against (both accept paths natively) and filters the SwiftFormat/Periphery reports after the fact, since Periphery has no per-path selector and always scans the whole module graph. Can also be passed as scope:/paths: key/value lines by a calling skill, along with an optional baseline: block (a prior per-rule-id issue count or list) used to report only the new-vs-baseline diff.

None — checks the whole package/project.

Outputs

  • A plain-text summary in the conversation — never a written file.

  • Findings grouped by tool (SwiftLint, the wired formatter, Periphery), then by rule id, with file:line and message.

  • Per tool: whether it ran or was skipped as "not wired"/"binary missing" (Step 1), a total issue count, and the severity breakdown where the tool has one.

  • When a baseline was supplied, only the new-vs-baseline diff per rule id/tool instead of the full current list.

  • No source file or SwiftLint/swift-format/.swiftformat/Periphery configuration is ever modified.

Execution flow

flowchart TD A["Start /iru-swift-code-quality\n(optional scope, or scope/paths/baseline key:value lines)"] --> B["Resolve project root:\nPackage.swift or .xcodeproj/.xcworkspace"] B --> C["Step 1: check which tools are wired\nSwiftLint (.swiftlint.yml + binary)\nformatter (.swift-format or .swiftformat + binary)\nPeriphery (.periphery.yml + binary)"] C --> D{"Config present but\nbinary missing?"} D -- yes --> E["AskUserQuestion:\ninstall via brew, or skip this tool"] D -- no --> F E --> F["Step 2: run each wired tool\n(scoped paths for SwiftLint/formatter,\nfull scan for Periphery)"] F --> G{"Underlying build\nfailed for a tool?"} G -- yes --> H["Report that tool's failure, stop for it\n(report not freshly regenerated)"] G -- no --> I["Step 3: parse each report,\nfilter to scope if given"] I --> J["Step 4: bucket by tool, then rule id;\nreport severity breakdown"] J --> K{"Baseline supplied?"} K -- yes --> L["Report only new-vs-baseline\nissues per tool/rule id"] K -- no --> M["Report full current totals\nand per-rule-id counts"]

Dependencies

Invokes

None — iru-swift-code-quality is a leaf skill; it only runs SwiftLint/swift format lint/SwiftFormat/Periphery and parses their reports, it does not call any other skill in this catalog.

Invoked by

  • Code — Step 3 and 7, quality baseline and comparison for Swift plans, via the iru-gate-runner agent.

  • Swift Code One Task Group — Steps 1 and 3, once per task group’s bucket, via the iru-gate-runner agent, to capture a pre-change baseline covering the bucket’s touched file(s) and later diff the post-change result against it.

Related agents

  • iru-gate-runner — both callers above always invoke this skill from inside a iru-gate-runner sub-agent they spawn, not inline.

Source

SKILL.md on GitHub — the file this page was generated from.