本文へ移動
cccskills
無料GitHub で公開

dependabot-config

Reference for `.github/dependabot.yml` - GitHub-native dependency-update orchestrator. Required keys (`version: 2`, `updates[]` array) plus per-update fields (`package-ecosystem`, `directory` / `directories`, `schedule.interval`); common optional fields (`ignore`, `groups`, `allow`, `labels`, `milestone`, `open-pull-requests-limit`, `target-branch`, `vendor`, `versioning-strategy`, `assignees`, `commit-message`); auto-rebase + grouped-PR + security-only updates. Use when authoring or reviewing Dependabot configs in GitHub-hosted repos.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qaスター 13(GitHub 13 + cccskills 0)
連携ツールGitHub

実行スコア 未検証

インストール方法を見る

含まれるファイル(13)

  • SKILL.md8.7 KB
  • .tessl-plugin/plugin.json674 B
  • evals/automerge-shipped-a-breaking-minor-at-2am/criteria.json5.9 KB
  • evals/automerge-shipped-a-breaking-minor-at-2am/task.md5.6 KB
  • evals/pr-flood-after-four-repos-became-one/criteria.json6.4 KB
  • evals/pr-flood-after-four-repos-became-one/task.md6.3 KB
  • evals/production-ships-from-a-branch-that-is-not-main/criteria.json5.0 KB
  • evals/production-ships-from-a-branch-that-is-not-main/task.md5.0 KB
  • evals/two-of-eight-manifests-ever-got-a-bump/criteria.json6.5 KB
  • evals/two-of-eight-manifests-ever-got-a-bump/task.md8.5 KB
  • evals/waiver-list-outlived-the-people-who-approved-it/criteria.json8.3 KB
  • evals/waiver-list-outlived-the-people-who-approved-it/task.md7.8 KB
  • references/dependabot-advanced.md2.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

dependabot-config

Overview

Per db-cfg, Dependabot opens PRs (or issues, for security-only updates) when a new version is available for a declared dependency, configured via .github/dependabot.yml at the repo root.

This is a reference skill - it defines the config surface; it doesn't run scans (that's snyk-test or osv-scanner). Dependabot complements SCA tools by automating the upgrade PR.

When to use

  • The repo is GitHub-hosted (Dependabot is GitHub-native).
  • Authoring a new .github/dependabot.yml.
  • Reviewing a .github/dependabot.yml PR for completeness + hygiene.
  • Migrating from manual update sweeps to automated orchestration.

For non-GitHub repos, see renovate-config.

Step 1 - Top-level structure

Per db-cfg:

version: 2
updates:
  - package-ecosystem: "npm"
    directory: "/"
    schedule:
      interval: "daily"

Required top-level keys:

KeyUse
versionAlways 2 (only supported version)
updatesArray of per-ecosystem update configurations

Step 2 - Per-update required fields

Per db-cfg:

FieldUse
package-ecosystemPackage manager: npm, bundler, cargo, composer, docker, github-actions, gitsubmodule, gomod, gradle, maven, mix, nuget, pip, pub, swift, terraform
directory (or directories)Manifest location relative to repo root
schedule.intervalFrequency: daily / weekly / monthly / quarterly / semiannually / yearly / cron (with cron expression)

directories (plural) supports an array for monorepos:

- package-ecosystem: "npm"
  directories:
    - "/services/api"
    - "/services/worker"
    - "/packages/shared"

Step 3 - Common optional fields

Per db-cfg:

ignore

"Ignore updates for dependencies with matching names, optionally using * to match zero or" more characters.

ignore:
  - dependency-name: "lodash"
    versions: [">=5.0.0"]    # don't update past v5
  - dependency-name: "*-internal-*"
    update-types: ["version-update:semver-major"]

groups

"Combines multiple dependency updates into single pull requests using pattern matching and dependency type filters."

groups:
  dev-deps:
    dependency-type: "development"
    update-types: ["minor", "patch"]
  production-deps:
    dependency-type: "production"
    exclude-patterns: ["express*", "fastify*"]

Grouped PRs reduce review noise - instead of 30 individual PRs for dev-deps, get one consolidated PR.

allow

"Restricts updates to explicitly listed dependencies only."

allow:
  - dependency-name: "react*"
  - dependency-type: "direct"

Use carefully - overly-narrow allow lists silently drop coverage of newly-added deps.

Other common fields

FieldUse
labelsCustom PR labels (overrides default dependencies)
milestoneNumeric milestone ID for created PRs
open-pull-requests-limitMax concurrent version PRs (default 5; security-only PRs not counted)
target-branchUpdate target branch (security PRs always go to default branch)
vendorMaintain vendored deps (Bundler, Go modules)
versioning-strategyauto / strict / increase-if-necessary / widen-ranges
assigneesGitHub usernames for assignment
commit-messageCustomize prefix + scope

Step 4 - Realistic multi-ecosystem example

One updates entry per ecosystem, each with its own schedule and labels (reuse the groups / ignore blocks from Step 3 as needed):

version: 2
updates:
  - package-ecosystem: "npm"
    directory: "/"
    schedule:
      interval: "weekly"
      day: "monday"
    open-pull-requests-limit: 10
    groups:
      dev-deps:
        dependency-type: "development"
    labels: ["dependencies", "javascript"]

  - package-ecosystem: "github-actions"
    directory: "/"
    schedule:
      interval: "monthly"

  - package-ecosystem: "docker"
    directory: "/Dockerfile"
    schedule:
      interval: "weekly"

Full config with per-ecosystem groups, ignore rules, assignees, and commit-message customization: references/dependabot-advanced.md.

Verify: after committing .github/dependabot.yml, open the repo's Dependabot view (Insights -> Dependency graph -> Dependabot) and confirm GitHub reports no dependabot.yml parse error and lists every configured package-ecosystem; if it flags a parse error, fix the reported key and re-push before relying on the config. Use the view's Check for updates control to force a run instead of waiting for the schedule.

Step 5 - Security updates (always-on)

Dependabot security updates are enabled separately in repo settings (Security → Code security and analysis → Dependabot security updates). Security PRs:

  • Always target the default branch (regardless of target-branch)
  • Don't count against open-pull-requests-limit
  • Trigger immediately on CVE disclosure (not on schedule)
  • Bypass ignore rules for the affected version (you can ignore the package generally, but Dependabot will still PR a security fix unless you explicitly ignore the CVE)

Step 6 - False-positive triage analogue

Dependabot doesn't produce findings to triage - it produces upgrade PRs. The "FP triage" analogue is suppressing unwanted update PRs:

MechanismUse
ignore.dependency-name + versions rangePin a dep to a major version (avoid breaking changes)
ignore.update-typesBlock all major-version PRs for a dep
Repo Settings → Security → Disable Dependabot for a specific packageCategorical disable (last resort)

Justification template (mandatory in dependabot.yml comments):

ignore:
  # Reason: react v19 + react-router v7 incompatibility blocks upgrade
  # Approved-by: alice@example.com
  # Re-review-date: 2026-09-15 (re-evaluate when react-router v8 ships)
  - dependency-name: "react"
    update-types: ["version-update:semver-major"]

Cadence: every quarter, audit ignore: entries; expired re-review dates removed.

Step 7 - Auto-merge integration

Dependabot creates PRs but doesn't auto-merge. For auto-merge, pair with GitHub Auto-merge or a workflow that reads dependabot/fetch-metadata and gates the merge to version-update:semver-patch only (after CI passes; patch-only is safer than minor / major). Full workflow: references/dependabot-advanced.md.

Anti-patterns

Anti-patternWhy it failsFix
interval: "daily" everywherePR storm overwhelms reviewersweekly for non-critical; daily only for security-sensitive deps
No groups: for dev depsEach dev-dep update is a separate PRGroup by dependency-type (Step 3)
ignore without expiration commentPermanent debtMandatory Re-review-date: (Step 6)
Skip security-only updates feature (or disable in Settings)Critical CVEs reach prodKeep enabled; never disable wholesale
Auto-merge minor/major automaticallyBreaking changes ship without reviewAuto-merge patch only (Step 7)

Limitations

  • GitHub-only - for GitLab / Bitbucket / non-GitHub, use renovate-config.
  • No support for non-language manifests (Helm Chart.yaml, pre-commit hook updates, etc.) without workarounds.
  • Grouped PRs introduce coupling - one breaking change blocks the whole group.
  • Per-CVE waivers (vs per-version-range) are not first-class; the ignore mechanism is dependency-centric.

References

  • db-cfg - official configuration reference
  • references/dependabot-advanced.md - full multi-ecosystem config example + patch-only auto-merge workflow
  • docs.github.com/en/code-security/dependabot - full Dependabot docs
  • docs.github.com/en/code-security/dependabot/dependabot-security-updates - security updates feature
  • github.com/dependabot/fetch-metadata - auto-merge helper
  • snyk-test, osv-scanner, renovate-config, npm-pip-maven-audit - sister tools

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

コメント

ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。

まだコメントはありません。最初のコメントを書いてみませんか。

送信するとすぐに公開されます。メールアドレスや電話番号などの個人情報は書かないでください。URL は 2 つまでです。

0 / 1000 文字

このフォームは reCAPTCHA で保護されており、Google のプライバシーポリシーと利用規約が適用されます。

同じリポジトリのスキル

Builds a CI gate that fails the build on **new** WCAG / a11y violations introduced by a PR while grandfathering pre-existing violations on a per-rule / per-page baseline. Aggregates verdicts from axe-core / pa11y / Lighthouse a11y / WAVE / IBM Equal Access scans. Use when a project has accumulated a11y debt and a strict "zero violations" gate would block every PR - the ratchet pattern lets the team ship while preventing regressions.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

Workflow skill that builds an A/B-test validity checklist from an experiment proposal, walking the canonical design-correctness gates - pre-registered OEC/power/guardrails, randomization unit + SRM check, assignment integrity, telemetry, peeking discipline, novelty/primacy, post-experiment SRM re-check - into a per-experiment checklist + sign-off form. Use when launching, auditing, or governing an experiment. For pitfall mechanics (guardrails, peeking) see experiment-results-interpreter's references; to read an already-valid result use experiment-results-interpreter; for per-SDK harness tests use experiment-sdk-testing - this gates DESIGN, not SDK code.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

The single install-flow skill: the reference contract for the PWA install flow (installability gate fields, `beforeinstallprompt` handshake, per-platform paths - Android WebAPK / iOS Share menu / Firefox no-op - and the `display-mode` post-install signal, in references/install-flow-reference.md) plus the build-an-X workflow that emits the Add-to-Home-Screen suite. Walks the four-stage timeline, emitting one test per gate cell per [web.dev/articles/install-criteria][install-criteria], the deferred-prompt → `prompt()` → `userChoice` chain per [web.dev/articles/customize-install][customize-install], the iOS Safari manual-metadata branch (`apple-touch-icon`) per [web.dev/learn/pwa/installation][learn-pwa], and the post-install `(display-mode: standalone)` MQ assertion. Output: a Playwright spec with per-stage cells plus a coverage matrix. Use when a PWA's manifest, icons, or install handler change, when install conversion drops at an unknown stage, or when triaging a flaky install assertion.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

Build-an-X workflow that uses an LLM to map existing tests to spec sections - given a spec doc + the test suite, the LLM identifies which tests cover which sections, surfaces uncovered sections (gap), and recommends specific tests to add. Output is a coverage matrix per spec ID. Scope is mapping tests that already exist and naming the gaps, not authoring tests for new acceptance criteria. Use when a spec doc and a test suite both exist but nobody can say which requirements are actually covered - before a release sign-off, an audit, or a decision about where to spend the next round of test effort.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

Generates tests from natural-language specs (acceptance criteria, user stories, requirements) using an LLM, with confidence scoring per test case (LLM self-assessment plus heuristics: assertion quality, naming, completeness), batching uncertain cases for human review, and integration with the team's existing test framework. Use when the user asks to generate unit tests from acceptance criteria, convert user stories to test cases, automate test creation from requirements, or augment a spec-driven test suite with AI-generated stubs that are then curated before merge.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

Configures Allure Report (test-runner adapter install, `allure-results` directory wiring, `categories.json` for failure classification, `history-trend.json` retention via the copy-history-between-runs pattern), runs the Allure CLI to convert `allure-results` to a static HTML site, and uploads the report as a CI artifact. Use when the team needs richer test reporting than JUnit XML - step-level attachments, per-test history, retry tracking, and severity / epic / feature labeling across framework-agnostic adapters (pytest, Jest, JUnit, TestNG, NUnit, Mocha). As a rich static HTML report generator, it is the open-source alternative to the sunset ExtentReports (JVM/.NET per-test HTML narrative); for hosted cross-run flakiness analytics rather than a static per-run report use currents-integration.

日本語の概要は準備中です。原文の説明を表示しています。

testland/qa132026年10月20日 更新

testland のスキルをすべて見る

このスキルの問題を報告する