Swift Generate All Tests

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

Bring an entire Swift codebase’s unit test coverage up to a minimum bar — 80% line coverage — in one pass: find every production type (struct, class, enum, actor) with no test at all or with measured coverage below the bar, then write or extend tests for each until it clears it. Swift Package Manager and Xcode projects only; it never runs against, or falls back to, any other language.

Purpose

iru-swift-generate-all-tests answers "bring this whole Swift codebase’s test coverage up to a bar" rather than "write tests for this one type/task," which is what iru-swift-code-one-task and the iru-code pipeline already handle per task. It explores the whole codebase to build an inventory of testable types, delegates to iru-swift-coverage for the exact, measured percentage per type (never an estimate), and only writes or extends test code — defaulting to Swift Testing (@Suite, @Test, #expect, #require) for unit tests and falling back to XCTest only for XCUITest and performance (measure) cases, following the project’s own test framework and style. For apps it also closes the UI-test gap under the catalog’s UI contract (see Building user interfaces): screens and components with no UI tests get component-capture tests and XCUITests that capture every .ui-quality.yml direction x theme and run an accessibility audit, then one full iru-swift-test ui run rebuilds .ui-screenshots/. It does not add license headers, DocC comments, or run a code-quality pass, and it only touches production source code when a type is genuinely untestable as written, which it reports explicitly rather than silently redesigning the type.

When .ui-quality.yml says accessibility: none (opted out entirely) or off (rules apply, checks not run), accessibility checks are not applicable: they are skipped and reported as "accessibility: opted out" (none) or "accessibility checks off" (off) — never "unverified" and never a failure.

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-generate-all-tests [scope]

Inputs

Argument Required Description Default

scope

No

A Swift package target or Xcode app target to limit the run to, or a comma-separated list of specific types within it. Accepted as a bare positional argument or as an args scope: <target-or-type-list> line.

The project’s primary library/app target

Outputs

  • New or extended test files for every type found with no matching test file, or with line coverage below 80% — defaulting to Swift Testing, following the project’s own test-file naming/source-root convention, and using XCTest only for XCUITest/measure cases.

  • Occasionally, a minimal testability change to a type under test (e.g. extracting a protocol, injecting a dependency) — only when a type is genuinely untestable as written, and always called out explicitly rather than folded silently into "tests added."

  • A report: total candidate types found (and any excluded as having no testable logic), which test framework(s) were detected/used, how many types already met the bar, before/after coverage per type touched, any type still below 80% after two extension attempts (with its exact remaining uncovered lines), and the full-suite test result.

  • For screens and components that had no UI tests: contract UI tests (component suites using UIScreenshots.render/capture, XCUITests with --fixture data and one performAccessibilityAudit() test), the shared UIScreenshots.swift helper when missing, and a report of screenshot counts per kind, malformed names, accessibility issues at the configured level, screens left uncovered (e.g. no UI-test target) and anything unverified.

  • No license headers, DocC comments, or code-quality fixes — those remain iru-check-license, iru-swift-docc, and `iru-swift-code-quality’s jobs.

Execution flow

flowchart TD A["Start /iru-swift-generate-all-tests [scope]"] --> B{"Package.swift or\n.xcodeproj/.xcworkspace found?"} B -- neither --> B1["Stop: Swift projects only"] B -- both found, scope unclear --> B2["AskUserQuestion:\npackage or app target?"] B2 --> C B -- yes --> C["Detect test framework(s):\nSwift Testing vs XCTest vs both"] C --> D["Explore agent: inventory every\nproduction type + existing test\nconvention + source root"] D --> E{"Test source root exists?"} E -- no --> E1["Stop: ask user, this skill\nextends tests, doesn't scaffold one"] E -- yes --> F["gate-runner: single swift-coverage run\nover every candidate type\n(batched if the list is large)"] F --> G["Classify: missing tests /\nbelow 80% / already sufficient"] G --> H["One agent per type (parallel batches):\nwrite/extend Swift Testing tests\n(XCTest only for XCUITest/measure)"] H --> H2["Step 5b: UI tests for screens and components\nwithout them (contract captures + accessibility\naudit; full iru-swift-test ui run)"] H2 --> I["Re-measure via swift-coverage,\nscoped to touched types"] I --> J{"Still below 80%?"} J -- yes, attempts left --> H J -- yes, 2 attempts used --> K["Report as residual gap\nwith exact uncovered lines"] J -- no --> L["gate-runner: run full suite\nonce (swift-test)"] K --> L L --> M{"Any failures?"} M -- yes, caused by this run --> N["Fix regression; re-run full suite"] N --> L M -- no / pre-existing --> O["Final report"]

Dependencies

Invokes

  • Swift Coverage — Step 3, and again in Step 6 to re-measure after generation: delegated through the iru-gate-runner agent so a large .profdata/llvm-cov/xccov report never lands directly in this conversation, and so the whole suite runs once per batch rather than once per type.

  • Swift Test — Step 5b (ui run) and Step 7, once, to confirm the full suite still passes after every type’s tests have been written or extended; also delegated through iru-gate-runner. Step 5b also runs it with the ui selector (a full UI run that clears and rebuilds .ui-screenshots/).

Invoked by

None from inside this catalog — it is a top-level entry point invoked by a person who wants coverage brought up across the whole Swift codebase in one pass, independent of the /iru-code//iru-issue pipeline (which handles per-task test-writing via iru-swift-code-one-task).

Related agents

  • Explore — Step 2, to build the type/test-file inventory across the whole codebase without reading every source file directly.

  • iru-gate-runner — Steps 3, 6, and 7, to run iru-swift-coverage/iru-swift-test and report back only a compact summary instead of a raw coverage/test log.

  • general-purpose (default, no subagent_type named) — Step 5, one per type needing new/extended tests, dispatched concurrently since each only touches its own test file; Step 5b reuses this for one agent per screen or component needing UI tests.

Source

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