flake-triage

flake-triage is a skill for Claude Code from yschimke/compose-ai-tools. It costs 70 tokens per session (1,469 once invoked), scanned A, original, Apache-2.0.

A method for checking whether a changed visual preview is a real code regression or an unstable render. It repeats the same preview at one commit and compares the resulting image files.

In plain words
What is it for?
Triage of visual-diff alerts, repeated preview rendering, comparing image outputs, and investigating nondeterministic screenshots, GIFs, or filmstrips.
Why use it?
Previews can change because of clocks, randomness, animation, or network images even when the relevant code did not change. Repeated renders show whether the preview is reproducible before anyone investigates the pull request.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is cd ../compose-preview-vscode.

Good fit Triage of visual-diff alerts, repeated preview rendering, comparing image outputs, and investigating nondeterministic screenshots, GIFs, or filmstrips.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/yschimke/compose-ai-tools
agentmods
npx agentmods add skills/yschimke/compose-ai-tools/flake-triage

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for flake-triage

README.md
[![agentmods](https://agentmods.dev/badge/skills/yschimke/compose-ai-tools/flake-triage.svg)](https://agentmods.dev/skills/yschimke/compose-ai-tools/flake-triage)
Your own site
<a href="https://agentmods.dev/skills/yschimke/compose-ai-tools/flake-triage"><img src="https://agentmods.dev/badge/skills/yschimke/compose-ai-tools/flake-triage.svg" alt="Measured on agentmods" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,469 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium MCP Rug Pull · line 50
    npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
    Fix: Pin the version: npx @scope/[email protected]
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00070 $0.01469
Opus 5 $0.00035 $0.00734
Sonnet 5 $0.00014 $0.00294
Haiku 4.5 $0.00007 $0.00147

Measured 8d ago against content hash 3263f346666e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

flake-triage scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/flake-triage/SKILL.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Is it a regression, or is the preview unstable?

A changed preview whose source the PR does not touch — clocks, timestamps, randomness, animation frames, network-loaded images — is instability to triage, not a regression to rubber-stamp and not a fix to write blind. The distinction is decidable in a few minutes, at one commit, with no reference to the base branch.

The oracle

Render the same preview N times at the same commit and compare the bytes. If two renders of one commit differ, nothing about the PR's diff is evidence of anything.

for i in 1 2 3; do
  ./gradlew :samples:android:composePreviewRender --rerun \
    -PcomposePreview.filter=<PreviewFunctionName>
  cp <module>/build/compose-previews/renders/<Preview>.png run-$i.png
done
md5sum run-*.png

--rerun is load-bearing: a render task goes UP-TO-DATE off its own outputs (including .error.json sidecars from a failed run), so a plain re-invocation compares a file against itself and always "passes".

A harness capture has the same oracle, and it is cheaper. The captures the visual-diff bot compares are Playwright shots of committed static fixtures, so N runs cost seconds and need no JVM.

The serve captures (serve-, viewer-) are not triaged here any more. That harness moved with the server to yschimke/compose-preview-server, where it runs as the visual-harness CI job; triage them in that repository, against its preview-harness/. What stays here is the fixture tree those captures read (preview-server/preview-harness/fixtures/, still asserted by ServeWebFixtureTest and friends) — a fixture edit here changes what that repository captures.

# compose-preview-vscode/preview-harness — the VS Code panel's own fixtures.
# Run from `compose-preview-vscode/`, not from inside the harness directory, and build
# the webview bundle first: the fixtures load it, so a stale one moves pixels for
# reasons no commit explains.
cd ../compose-preview-vscode
node esbuild.webview.mjs
HARNESS_CHROMIUM=/path/to/chromium HARNESS_FIXTURE=<fixture> HARNESS_THEME=light \
  npx playwright test -c preview-harness/playwright.config.mjs snapshot.spec.mjs
md5sum preview-harness/out/<capture>.light.png

Read the full file on GitHub · 118 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 118 lines · 70 tokens per session scan A 3263f346666e

Subscribe to this mod's changes

flake-triage is a skill published in the GitHub repository yschimke/compose-ai-tools (110 stars, last pushed today), licensed Apache-2.0. It adds 70 tokens to every session and 1,469 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

flutter-skill

Automate and test Flutter applications — launch apps, inspect widgets, tap elements, enter text, scroll, swipe, take screenshots, validate state, and debug via Dart VM Service Protocol. Use when the user wants to run Flutter app tests, automate Flutter UI interactions, inspect widget trees, debug a running Flutter…

ai-dashboad/flutter-skill · 73 tokens

roborazzi

Use when working with Roborazzi screenshot tests on Android/JVM — setting up the Roborazzi Gradle plugin, running record/compare/verify tasks, writing tests with captureRoboImage or RoborazziRule, Compose Preview screenshot testing (ComposablePreviewScanner), Compose Multiplatform (iOS/desktop) screenshots, AI-powered…

takahirom/roborazzi · 95 tokens

running-bug-review-board

Runs real-user QA, manual test plans, UX bug hunts, build sign-off, bug filing, and bug triage for web or iOS/iPadOS apps. Use when asked "QA this", "is this ready to ship?", or similar. Produces P0/P1/P2 bug reports, YES/NO phase sign-off, tracker sync guidance, and an HTML QA dashboard; keeps Interactive BRB triage…

RayFernando1337/rayfernando-skills · 96 tokens

ios-testing

Testing patterns for Swift and SwiftUI apps.

TalissonVitorino/kmp-ios-skills · 69 tokens

mobile-testing

Android and JVM testing - JUnit5, MockK, Turbine for Flow, and Compose UI testing for unit, integration, and UI tests; also applies to KMP commonTest running on the JVM/Android target. For iOS/Swift tests (XCTest, Swift Testing, XCUITest) use ios-testing. For TDD methodology and the three-tier test model (fake-first…

TalissonVitorino/kmp-ios-skills · 102 tokens

mushi-integration

Full end-to-end Mushi Mushi integration smoke test: bug capture → AI triage → story mapping → TDD test generation → approval → execution → PDCA cycle. Use when "test mushi integration", "verify full pipeline", "mushi e2e check", "does mushi work end-to-end", "smoke test mushi", or after deploying changes.

kensaurus/cursor-kenji · 82 tokens