← Build macOS AppsCONTENT HISTORYWHAT CHANGED · RULE-BASED ANALYSIS
Update to Build macOS Apps
Snapshot Sep 30, 2026 · 23:18 UTC · version 0.1.4
Collection source: not recorded for this historical snapshot.
First saved snapshot
No earlier snapshot is available to establish a change.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"name": "test-triage",
"description": "Triage macOS tests across Xcode and SwiftPM. Use when narrowing failures, explaining assertions or crashes, or separating setup from regressions.",
"included_files": [
{
"relative_path": "agents/openai.yaml",
"size_in_bytes": 283
}
],
"skill_md_contents": "---\nname: test-triage\ndescription: Triage macOS tests across Xcode and SwiftPM. Use when narrowing failures, explaining assertions or crashes, or separating setup from regressions.\n---\n\n# Test Triage\n\n## Quick Start\n\nUse this skill to run the smallest meaningful test scope first, classify\nfailures precisely, and avoid treating every test failure like a product bug.\n\n## Workflow\n\n1. Detect the test harness.\n - Use `xcodebuild test` for Xcode-based projects.\n - Use `swift test` for SwiftPM packages.\n\n2. Narrow the scope.\n - If the user gave a target, product, or test filter, use it.\n - If not, prefer the smallest likely failing target before a full suite.\n\n3. Classify the result.\n - Build failure\n - Assertion failure\n - Crash or signal\n - Async timing or flake\n - Environment or fixture setup issue\n - Missing entitlement or host app issue\n\n4. Rerun intelligently.\n - Use focused reruns when a specific case fails.\n - Avoid burning time on full-suite reruns without new information.\n\n5. Summarize clearly.\n - What command ran\n - Which tests failed\n - What kind of failure it was\n - The best next proof step or fix path\n\n## Guardrails\n\n- Distinguish compilation failures from test execution failures.\n- Call out when a test appears to assume iOS-only or simulator-only behavior.\n- Mark likely flakes as such instead of overstating confidence.\n\n## Output Expectations\n\nProvide:\n- the command used\n- the smallest failing scope\n- the top failure category\n- a concise explanation of the likely cause\n- the next rerun or fix step\n"
}SHA-256: d808db42f464e4a9e791ba448ebd25f81e7dd032941ca278760cdae4ef0b6694