ci-flake-hunting

ci-flake-hunting is a skill for Claude Code from skyfox675/agents-skills. It costs 293 tokens per session (4,578 once invoked), scanned A, original, MIT.

Guidance for finding, explaining, and fixing flaky tests in continuous integration, the automated checks that run when code changes are proposed. A flaky test fails intermittently even when the code has not changed.

In plain words
What is it for?
Use it to investigate intermittent CI failures, trace their causes, apply fixes that remove instability, and improve the reliability of merge-queue builds.
Why use it?
It focuses on identifying instability that can wrongly reject correct changes and disrupt an automated merge queue.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents.

Good fit Use it to investigate intermittent CI failures, trace their causes, apply fixes that remove instability, and improve the reliability of merge-queue builds.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/skyfox675/agents-skills/ci-flake-hunting
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add skyfox675/agents-skills --skill ci-flake-hunting
Clone the repo
git clone --depth 1 https://github.com/skyfox675/agents-skills

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 ci-flake-hunting

README.md
[![agentmods](https://agentmods.dev/badge/skills/skyfox675/agents-skills/ci-flake-hunting/github.svg)](https://agentmods.dev/skills/skyfox675/agents-skills/ci-flake-hunting)
Your own site
<a href="https://agentmods.dev/skills/skyfox675/agents-skills/ci-flake-hunting"><img src="https://agentmods.dev/badge/skills/skyfox675/agents-skills/ci-flake-hunting/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ci-flake-hunting

Your own site · 80×15
<a href="https://agentmods.dev/skills/skyfox675/agents-skills/ci-flake-hunting"><img src="https://agentmods.dev/badge/skills/skyfox675/agents-skills/ci-flake-hunting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 293 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,578 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 pass 7 Sept 2026
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 $0.00293 $0.04578
Opus 5 $0.00147 $0.02289
Sonnet 5 $0.00059 $0.00916
Haiku 4.5 $0.00029 $0.00458

Measured 12d ago against content hash 0979bf41139d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

ci-flake-hunting 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 12d 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/ci-flake-hunting/SKILL.md · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CI Flake Hunting (authoritative deflake method)

A serial merge queue ejects the correct PR at the front the moment a flaky speculative build reds, AND disarms its auto-merge — the PR drops to a green-but-unqueued state and sits idle. So every live flake is a tax on every PR behind it. Killing one high-frequency flake measurably raises queue throughput; this is among the highest-leverage CI work there is.

This skill pairs with the ci-speed-hunting skill and the test-driven-development skill where available (whether the resulting test is well-shaped), and builds on the merge-queue mechanics in driving-prs-to-merge. Where a "make it pass" instinct conflicts with this skill's fix-forward rule, this skill wins.

Lane — what you own, and the two lanes you are constantly confused with

Read this before anything else. Getting it backwards is the one mistake that strands a broken queue.

You own the systemic hunt for instability. Concretely, these are YOURS:

  • merge_group / queue failures and the ejections they cause — the speculative build reds, the correct PR at the front is ejected and its auto-merge disarmed. Killing that flake is the highest-leverage work there is.
  • Post-merge failures on <integration-branch> — a red that only shows up after the merge lands.
  • The recurring flake itself — a spec that fails then passes on retry with no code change, mined across CI history and killed permanently.
  • Transient-infra reds — classify (Step 0), re-run, open NO code PR. Classifying a red as infra is doing your job, not handing it away.

Two lanes you must not absorb, and must not hand your work to:

  1. pr-checks (the PR watcher) owns a given open PR's failing checks — not you. When one PR's gate is red, that watcher unblocks that PR (re-run a transient, dispatch a fix for a real diff failure, dequeue and draft a deterministic poison pill, resolve conflicts). You do not babysit individual PRs' red checks. You may freely mine those failures as evidence to find a recurring flake — reading them is how you hunt — but owning the PR's route to green is theirs.

Read the full file on GitHub · 205 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 12d ago First seen · 205 lines · 293 tokens per session scan A 0979bf41139d

Subscribe to this mod's changes

ci-flake-hunting is a skill published in the GitHub repository skyfox675/agents-skills (11 stars, last pushed 12d ago), licensed MIT. It adds 293 tokens to every session and 4,578 once invoked, about $0.0015 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 2026-08-31.

Related

Other skills, from other repositories

Verification & Quality Assurance

Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.

ruvnet/ruflo · 36 tokens

ci

Configure Ginkgo for continuous integration — the recommended CLI flag set and the rationale for each flag (-r -p --randomize-all --randomize-suites --fail-on-pending --fail-on-empty --keep-going --cover --race --trace --json-report --timeout --poll-progress-after/-interval), invoking via go run to pin the CLI to…

onsi/ginkgo · 151 tokens

migrate-vstest-to-mtp

Use this skill before answering, planning, or editing whenever .NET tests or CI are switching from VSTest to Microsoft.Testing.Platform (MTP), or an MTP migration behaves differently. Triggers include "switch from VSTest"; MSTest/NUnit/xUnit MTP enablement; OutputType=Exe only for test projects in…

dotnet/skills · 191 tokens

playwright-ci

Production-ready CI/CD configurations for Playwright — GitHub Actions, GitLab CI, CircleCI, Azure DevOps, Jenkins, Docker, parallel sharding, reporting, code coverage, and global setup/teardown.

zebbern/claude-code-guide · 48 tokens

ci-maintenance-workflow

CI and GitHub Actions maintenance workflows — fix a failing test from a CI URL, fix a failing smoke test, add @pytest.mark.slow markers to slow tests, or review a PR against agent-checkable standards. Use when user asks to fix a failing test, fix a smoke test, mark slow tests, or review a PR. Trigger when the user…

UKGovernmentBEIS/inspect_evals · 120 tokens

playwright-testing

E2E testing with Playwright - Page Objects, cross-browser, CI/CD.

alinaqi/maggy · 20 tokens