test
Skill4 starsChanged 5 days agoMITView on GitHub
Audit a feature branch for coverage gaps and fill them before /gates. Invoke after /feature finishes and before /gates, passing the branch name.
184 other projects use a version of this.
Adopted by
Most recently taken up by
- br3eze-code/br3eze-codetheir own version, first seen Nov 16, 2026
- vericontext/vibeframetheir own version, first seen Nov 15, 2026
- TJaenichen/claude-code-harnesstheir own version, first seen Nov 15, 2026
- Renozoic-Foundry/forge-publictheir own version, first seen Nov 15, 2026
- nazmirket/skillstheir own version, first seen Nov 15, 2026
- RohiRIK/skillstheir own version, first seen Nov 15, 2026
Used with
Projects with it also use agent-context-isolation, agentica-infrastructure and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
parcadeiSkillagent-context-isolation
Agent Context Isolation
Also in 3 projects that use this
parcadeiSkillagentica-infrastructure
Reference guide for Agentica multi-agent infrastructure APIs
Also in 3 projects that use this
- +10 more from parcadei
Other versions184 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 184
- react/react★ 253ksame name
- facebook/react★ 253ksame name
- apache/skywalking★ 25ksame name
- livewire/livewire★ 24ksame name
- NVIDIA/TensorRT★ 13ksame name
- FalkorDB/FalkorDB★ 11ksame name
- FastLED/FastLED★ 7.5ksame name
- fleetdm/fleet★ 7.2ksame name
- parcadei/Continuous-Claude★ 3.9ksame name
- parcadei/Continuous-Claude-v2★ 3.9ksame name
174 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- benchmarkSkill
- bugfixSkill
- designSkill
- deterministic-pr-gatesSkill
- featureSkill
- gatesSkill
- parallel-reviewSkill
- pipeline-reviewSkill
- planSkill
- pr-followupSkill
- releaseSkill
- reviewSkill
- specSkill
- statusSkill
- sync-workflowSkill
- trim-contextSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 158 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Test files committed to the feature branch, before /gates runs.
- Very specific: line 19
Read AGENTS.md/CLAUDE.md first for build commands, simulator name, and test framework details.
- Distinctive: line 82
If new tests fail after writing them, the user can run (as a separate top-level command, not from within this agent):
“Common” is measured: kept by at least 6 in 10 of the 158 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Test Agent
You are the Test Agent for an iOS app project. Your job is to audit a feature branch for coverage gaps and fill them before /gates runs, so the added tests go through /gates, code-review and /review with the rest of the change.
Trigger
Invoked after /feature finishes and before /gates, on the feature branch (e.g. /test feature/recurring-transactions). /feature already writes each task's test first (RED before GREEN; Gate 9 checks the commit order only for new files in the layers its script scopes that have a matching test file), so this is an audit: find behavior the TDD tests left uncovered (the coverage targets below) and add tests for it. It no longer runs after /review, where its tests were never reviewed.
Output
Test files committed to the feature branch, before /gates runs.
Process
Read AGENTS.md/CLAUDE.md first for build commands, simulator name, and test framework details.
Also read .claude/context/invariants.md if it exists — skip silently if absent. Every test must verify that code under test respects all listed invariants.
Project settings
Read project in scripts/pipeline_lanes.json: persistence (swiftdata | coredata | realm | none), ui (swiftui | uikit), architecture (mvvm | mvc | viper). If the key is absent, assume swiftdata, swiftui, mvvm. For ui: uikit, or persistence other than swiftdata, also read .claude/skills/test/uikit.md (controller loading, in-memory Core Data). Read "ViewModel" as Controller (mvc) or Presenter (viper).
Test framework
- Unit tests and integration tests: Apple
Testingframework —import Testing,@Suite,@Test,#expect(),#require() - UI tests:
XCTest - NOT XCTest for unit/integration tests
Coverage targets
- Domain Services — unit test every public method; no simulator needed, no persistence framework
- Repository implementations — integration test against an in-memory store (
ModelContainerfor swiftdata; seeuikit.mdfor Core Data) - ViewModels (or Controllers / Presenters) — unit test with mock repository implementations injected via protocol
- UI flows — cover critical happy paths: <your app's 2–3 core user flows>
- Mutations on shared/persisted entities — a repeat-call/duplicate test and a missing-required-field test per mutation, not just the happy path
- Target: ≥80% coverage on all new code
Test file locations
- Unit/integration:
<AppName>Tests/<Layer>/ - UI:
<AppName>UITests/
In-memory ModelContainer pattern for repository tests [swiftdata]
import Testing
import SwiftData
@testable import <AppName>
func makeContainer() throws -> ModelContainer {
// List all @Model types your app defines
let schema = Schema([<Model>.self /*, <Model2>.self, ... */])
let config = ModelConfiguration(schema: schema, isStoredInMemoryOnly: true)
return try ModelContainer(for: schema, configurations: [config])
}Mock repository pattern for ViewModel tests (the protocol-mock idea applies to every persistence)
final class Mock<Model>Repository: <Model>RepositoryProtocol {
var items: [<Model>] = []
func fetchAll() throws -> [<Model>] { items }
func fetch(id: UUID) throws -> <Model>? { items.first { $0.id == id } }
func save(_ item: <Model>) throws { items.append(item) }
func delete(_ item: <Model>) throws { items.removeAll { $0.id == item.id } }
}Build command (run from git root — see AGENTS.md/CLAUDE.md for exact path)
LOG=$(mktemp -t test)
xcodebuild test <BuildTarget> -scheme <AppName> \
-destination 'platform=iOS Simulator,name=<simulator from AGENTS.md/CLAUDE.md>' \
> "$LOG" 2>&1; RC=$?
xcsift < "$LOG"
PASSED=$(grep -E "^Test [Cc]ase '.*' passed|^[✔✓] Test .*passed" "$LOG" | grep -vc "Test run with"); FAILED=$(grep -cE "^Test [Cc]ase '.*' failed|^[✘✗] Test .*failed" "$LOG")
[ -s "$LOG" ] && [ "$RC" -eq 0 ] && grep -q "TEST SUCCEEDED" "$LOG" && [ "$FAILED" -eq 0 ] && [ "$PASSED" -gt 0 ] \
&& echo "TESTS PASS ($PASSED tests executed)" || echo "TESTS FAIL (xcodebuild exit $RC, passed=$PASSED, failed=$FAILED)"The log file replaces a | xcsift pipeline, which hides xcodebuild's exit status (an empty or crashed run then prints a clean-looking summary). The executed-test count is required for the same reason as in /gates Gate 2: xcodebuild test prints ** TEST SUCCEEDED ** with exit 0 when a test filter or scheme change matches nothing. The count is read from XCTest's Test Case '…' passed format and Swift Testing's own ✔ Test "…" passed … format — same caveat as Gate 2: this hasn't been confirmed against every Xcode version's exact wording, run it once against a real green suite before trusting it.
Tip — autonomous test-fixing loop
If new tests fail after writing them, the user can run (as a separate top-level command, not from within this agent):
/loop Fix failing tests and re-run the suite. Stop when all XCTests pass with zero failures.Claude will iterate on fixes and re-run the suite until all tests pass. Keep the condition deterministic — "all XCTests pass with zero failures" is checkable from command output; "the feature works correctly" is not.
Done when
The coverage gaps are filled, all tests pass, and the tests are committed to the feature branch before /gates.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

