Agent skill

Check Cross Runtime

by ayutaz in ayutaz/piper-plus

Python canonical (src/pythonrun/piperplus/, src/python/pipertrain/, src/python/g2p/piperplusg2p/) を変更した PR で、 ONNX I/O 以外の追随漏れ (phonemizer / config schema / CLI flag / data 形式 / API 変更) を 7 ランタイム +…

MITAuto-check passedDevOps & Cloud

Install Check Cross Runtime

skills CLI
$ npx skills add ayutaz/piper-plus --skill check-cross-runtime -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ayutaz/piper-plus check-cross-runtime --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ayutaz/piper-plus.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/check-cross-runtime .claude/skills/check-cross-runtime && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
check-cross-runtime
GitHub stars
230
Token cost
~3.5k tokens
SKILL.md length
850 words
Files
1
Skills in repo
20
Repo updated
First seen
Licence
MIT

At a glance

Python canonical (src/pythonrun/piperplus/, src/python/pipertrain/, src/python/g2p/piperplusg2p/) を変更した PR で、 ONNX I/O 以外の追随漏れ (phonemizer / config schema / CLI flag / data 形式 / API 変更) を 7 ランタイム +…

  • Works in 7 steps: canonical Python が触られているか確認 → ONNX グラフ I/O 変更なら別 skill に委譲 → 7 ランタイム + 2 docker の touch 状況を一覧表示 → …
  • Tasks that involve Containers
  • SKILL.md covers なぜこの skill が必要か (既存 2 skill…, 過去事故サマリ, スコープ判定 (何を見るか) and 影響度の 3 段階分類, plus 5 more sections
  • Calls uv, git and gh

What it does

Check Cross Runtime is an agent skill from ayutaz/piper-plus. Python canonical (src/pythonrun/piperplus/, src/python/pipertrain/, src/python/g2p/piperplusg2p/) を変更した PR で、 ONNX I/O 以外の追随漏れ (phonemizer / config schema / CLI flag / data 形式 / API 変更) を 7 ランタイム + 2 docker image 範囲で fail-fast 検出する。PR

Its SKILL.md is about 3.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in DevOps & Cloud, covering Containers. It works with Python, Docker, ONNX and C++. The repository describes itself as: Multilingual neural TTS (6 languages: JA/EN/ZH/ES/FR/PT, code supports SV) — C++, C, Rust, Go, Python, npm (WASM). VITS + Prosody, streaming, CUDA/CoreML/DirectML. pip install…. The licence is MIT.

When your agent uses it

  • Tasks that involve Containers

Example prompts

  • “/check-cross-runtime”

Requirements

  • Python 3
  • Docker
  • Pre-approved tools (allowed-tools): Bash(git diff *), Bash(git status *), Bash(grep *), Bash(ls *), Bash(python *), Bash(uv run *)

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. canonical Python が触られているか確認
  2. ONNX グラフ I/O 変更なら別 skill に委譲
  3. 7 ランタイム + 2 docker の touch 状況を一覧表示
  4. カテゴリ別 grep ヒント (詳細調査)
  5. CLI / config / schema の version field 変更チェック
  6. 既存 sync gate を手動実行 (影響度 Medium 以上)
  7. Docker image の追随確認 (PR #443 の盲点)

What it can do on your machine

Read from SKILL.md and the folder at commit 421d066. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Bash(git diff *)
    • Bash(git status *)
    • Bash(grep *)
    • Bash(ls *)
    • Bash(python *)
    • Bash(uv run *)

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • uv
    • git
    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use uv, git and gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Check Cross Runtime loads about 3.5k tokens when it runs. Until then it costs about 66 tokens; SKILL.md has 850 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~66
When it runs · the whole SKILL.md, loaded when a task matches
~3.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ayutaz/piper-plus at commit 421d066, republished under its MIT licence (© ayutaz). 850 words, ~3,520 tokens.

Download SKILL.mdSave it as .claude/skills/check-cross-runtime/SKILL.md (or your agent's skills folder).
name
check-cross-runtime
description
Python canonical (`src/python_run/piper_plus/`, `src/python/piper_train/`, `src/python/g2p/piper_plus_g2p/`) を変更した PR で、 ONNX I/O 以外の追随漏れ (phonemizer / config schema / CLI flag / data 形式 / API 変更) を 7 ランタイム + 2 docker image 範囲で fail-fast 検出する。PR
allowed-tools
Bash(git diff *), Bash(git status *), Bash(grep *), Bash(ls *), Bash(python *), Bash(uv run *)
disable-model-invocation
true
<!-- editorconfig-checker-disable-file -->

Python 変更 → N ランタイム追随漏れチェック

src/python_run/piper_plus/*.py、 src/python/piper_train/*.py、 または src/python/g2p/piper_plus_g2p/*.py を変更した PR で、 他 7 ランタイム (C# / Rust × 2 crate / Go / JS-WASM / C API / iOS SPM / Kotlin-Android / Swift G2P) + 2 docker image (python-inference / webui) が 同じ振る舞いに 追随しているか を PR 提出前に確認する。

なぜこの skill が必要か (既存 2 skill との関係)

skillスコープ補完範囲
/check-runtime-parityONNX グラフ I/O (入出力名・形状・dtype) の変更推論バックエンドだけ
/check-new-runtime-asset新規データファイル (JSON/TOML/TSV) の 7 箇所 metadata 配布wheel/sdist 同梱漏れだけ
/check-cross-runtime (本 skill)それ以外の Python 変更 = phonemizer ロジック / config field / CLI flag / data file 形式変更 / API 変更 / docker recipe既存 2 skill の網からこぼれる広い領域

3 skill を組み合わせると、 Python canonical の変更が 7 ランタイム + 2 docker 全部に伝播しているかを 1 PR ごとにチェックできる。

過去事故サマリ

PRPython 修正内容追随漏れ検出方法
#391speaker_embedding 形状修正5 ランタイム (Rust / Go / C# / C++ / WASM) が古い形状のままPR #443 で偶然発見 (silent regression)
#443MB-iSTFT speaker_embedding 未対応の修正4 runtime (Rust / Go / C# / WASM) + 2 docker image (python-inference / webui) で同じバグ手動レビューでようやく発見
#384ZH-EN loanword 辞書 schema 拡張Rust / Go / C# / WASM / C++ で旧 schema のままパース失敗CI gate を後付けで作成

共通パターン:

  1. Python で 1 関数 / 1 schema を 「サクッと」 修正
  2. ローカル pytest は green、 CI も Python 観点だけ通る
  3. レビュアーも Python diff しか見ない
  4. 別 PR / リリース後ユーザー報告で 「他ランタイム壊れている」 が判明

この skill は PR 提出前 に local で同じ修正が他ランタイムにも入っているか チェックリスト化する。

スコープ判定 (何を見るか)

Canonical Python の 3 source of truth
カテゴリパス対応する他ランタイム
A. ランタイム推論src/python_run/piper_plus/*.py (voice.py, http_server.py, text_splitter.py, timing.py, model_manager.py)C# / Rust core / Go / WASM / C API / iOS SPM
B. G2P / phonemizersrc/python/g2p/piper_plus_g2p/*.py (japanese.py, chinese.py, multilingual.py, ssml.py 等)Rust piper-plus-g2p / Go phonemize / C# Phonemize / WASM g2p / C++ *_phonemize.cpp / Kotlin / Swift G2P
C. モデルエクスポート / 学習src/python/piper_train/*.py (export_onnx.py / vits/models.py / infer_onnx.py)ONNX I/O 変更なら /check-runtime-parity を使う。 それ以外の周辺ロジック (chunking / scaling / sentence split 等) が本 skill 対象
スコープ外 (本 skill では skip)
  • ONNX 入出力名 / 形状 / dtype の追加・改名 → /check-runtime-parity
  • 新規 JSON/TOML データファイル追加 → /check-new-runtime-asset
  • ruff format / type stub のみ → 何もしない
  • tests/, docs/, .github/workflows/, CHANGELOG.md のみ → 何もしない

影響度の 3 段階分類

影響度例推奨アクション
軽微 (Low)内部リファクタ・docstring・型注釈・テスト追加他 runtime touch 不要、 PR description に明記して skip
中 (Medium)新規 phonemizer 関数追加 / 既存関数の挙動拡張 / 新規 CLI flag / 公開 API への field 追加Rust + Go + C# を同 PR で実装、 WASM/C++/iOS/Kotlin/Swift は follow-up issue 化
重大 (High)config schema 変更 / data file 形式変更 / model schema 変更 / 互換性破壊全 7 runtime + 2 docker 同 PR が望ましい。 不可なら release blocker issue を立ててから merge

判定の目安:

  • 「他 runtime に同じバグが潜むか?」 → Yes なら Medium 以上
  • 「version field を bump したか?」 → Yes なら High
  • 「ユーザーが直接触る面 (CLI / HTTP / library API) か?」 → Yes なら Medium 以上

実行手順

1. canonical Python が触られているか確認
bash
CANONICAL_PATTERN='^(src/python_run/piper_plus/|src/python/piper_train/|src/python/g2p/piper_plus_g2p/).*\.py$'

CANONICAL_TOUCHED=$(git diff --name-only origin/dev...HEAD | grep -E "$CANONICAL_PATTERN" || true)

if [ -z "$CANONICAL_TOUCHED" ]; then
  echo "canonical Python は変更されていない — このチェックは skip 可"
  exit 0
fi

echo "=== Canonical Python 変更検出 ==="
echo "$CANONICAL_TOUCHED"
2. ONNX グラフ I/O 変更なら別 skill に委譲
bash
# /check-runtime-parity が担う領域は skip して責務分離
ONNX_IO_CHANGE=$(git diff origin/dev...HEAD -- \
  src/python/piper_train/export_onnx.py \
  src/python/piper_train/vits/models.py | \
  grep -E '^[+-].*(speaker_embedding|prosody_features|language_id|input_names|output_names)' || true)

if [ -n "$ONNX_IO_CHANGE" ]; then
  echo "WARNING: ONNX I/O 変更を検出。 /check-runtime-parity も併用すること"
fi
3. 7 ランタイム + 2 docker の touch 状況を一覧表示
bash
declare -A RUNTIME_PATHS=(
  ["Python runtime"]="src/python_run/piper_plus/"
  ["C# Core"]="src/csharp/PiperPlus.Core/"
  ["C# CLI"]="src/csharp/PiperPlus.Cli/"
  ["Rust core"]="src/rust/piper-core/"
  ["Rust CLI"]="src/rust/piper-cli/"
  ["Rust G2P"]="src/rust/piper-plus-g2p/"
  ["Rust WASM"]="src/rust/piper-wasm/"
  ["Go runtime"]="src/go/piperplus/"
  ["Go CLI"]="src/go/cmd/piper-plus/"
  ["Go G2P"]="src/go/phonemize/"
  ["WASM npm (TTS)"]="src/wasm/openjtalk-web/"
  ["WASM npm (G2P)"]="src/wasm/g2p/"
  ["C API"]="src/cpp/"
  ["iOS SPM"]="Sources/PiperPlus/"
  ["Swift G2P"]="Sources/PiperPlusG2P/"
  ["Kotlin Android"]="android/piper-plus-g2p/"
  ["Docker python-inference"]="docker/python-inference/"
  ["Docker webui"]="docker/webui/"
)

echo ""
echo "=== 7 runtime + 2 docker の touch 状況 ==="
CHANGED_FILES=$(git diff --name-only origin/dev...HEAD)
for label in "${!RUNTIME_PATHS[@]}"; do
  path="${RUNTIME_PATHS[$label]}"
  if echo "$CHANGED_FILES" | grep -q "^${path}"; then
    echo "  [TOUCHED  ] $label  ($path)"
  else
    echo "  [UNTOUCHED] $label  ($path)  <-- 追随確認"
  fi
done
4. カテゴリ別 grep ヒント (詳細調査)
A. ランタイム推論 (src/python_run/piper_plus/ を変更したとき)
bash
# 変更された関数名を抽出
git diff origin/dev...HEAD -- src/python_run/piper_plus/voice.py | \
  grep -E '^[+-][[:space:]]*def ' | sort -u

# 同名 / 類似シグネチャを他 runtime で grep
SYMBOL='synthesize'  # ← 上で見つけたシンボル名で置き換え
echo "=== Rust ==="; grep -rn "$SYMBOL" src/rust/piper-core/src/ 2>/dev/null | head
echo "=== Go ==="; grep -rn "$SYMBOL" src/go/piperplus/ 2>/dev/null | head
echo "=== C# ==="; grep -rn "$SYMBOL" src/csharp/PiperPlus.Core/ 2>/dev/null | head
echo "=== WASM ==="; grep -rn "$SYMBOL" src/wasm/openjtalk-web/src/ 2>/dev/null | head
echo "=== C++ ==="; grep -rn "$SYMBOL" src/cpp/ 2>/dev/null | head
B. G2P / phonemizer (src/python/g2p/piper_plus_g2p/ を変更したとき)
bash
# 変更された言語ファイル名を取得
G2P_FILES=$(git diff --name-only origin/dev...HEAD | \
  grep -E '^src/python/g2p/piper_plus_g2p/[a-z_]+\.py$' | \
  xargs -I{} basename {} .py)

for lang in $G2P_FILES; do
  echo "=== $lang phonemizer の 8 runtime 同期確認 ==="
  echo "  Python:    src/python/g2p/piper_plus_g2p/${lang}.py  [canonical]"
  ls -la "src/rust/piper-plus-g2p/src/${lang}.rs"           2>/dev/null && echo "  Rust:      OK" || echo "  Rust:      MISSING"
  ls -la "src/go/phonemize/${lang}.go"                       2>/dev/null && echo "  Go:        OK" || echo "  Go:        MISSING"
  ls -la "src/csharp/PiperPlus.Core/Phonemize/"              2>/dev/null | grep -i "$lang" || echo "  C#:        要確認"
  ls -la "src/wasm/g2p/src/${lang}"                          2>/dev/null && echo "  WASM:      OK" || echo "  WASM:      MISSING"
  ls -la "src/cpp/${lang}_phonemize.cpp"                     2>/dev/null && echo "  C++:       OK" || echo "  C++:       MISSING"
  ls -la "android/piper-plus-g2p/src/main/"                  2>/dev/null | grep -i "$lang" || echo "  Kotlin:    要確認"
  ls -la "Sources/PiperPlusG2P/"                             2>/dev/null | grep -i "$lang" || echo "  Swift:     要確認"
done
C. モデルエクスポート / 学習周辺 (src/python/piper_train/ の非 ONNX-I/O 変更)
bash
# 例: infer_onnx.py の chunking / scaling ロジックを変更したとき、
# 同じロジックが他 runtime にも存在するかを確認
git diff origin/dev...HEAD -- src/python/piper_train/infer_onnx.py | \
  grep -E '^\+' | grep -iE '(scale|chunk|silence|split|trim)' | head
5. CLI / config / schema の version field 変更チェック

docs/spec/*.toml 内の version / schema_version を bump したかを確認:

bash
SPEC_VERSION_CHANGE=$(git diff origin/dev...HEAD -- 'docs/spec/*.toml' | \
  grep -E '^[+-].*(version|schema_version) *=' || true)

if [ -n "$SPEC_VERSION_CHANGE" ]; then
  echo "=== Spec version 変更検出 — 影響度 High の可能性 ==="
  echo "$SPEC_VERSION_CHANGE"
  echo ""
  echo "対応する sync gate を必ず手動実行:"
  echo "  uv run python scripts/check_phoneme_set_version.py"
  echo "  uv run python scripts/check_language_id_map_contract.py"
  echo "  uv run python scripts/check_dictionary_versions.py"
fi
6. 既存 sync gate を手動実行 (影響度 Medium 以上)

該当する canonical 変更ごとに、 対応する sync gate を PR 提出前に local で 通しておく (CI で気付くと cycle of shame):

bash
# 全 sync gate を一気に回す (推奨)
uv run python scripts/check_loanword_consistency.py
uv run python scripts/check_loanword_forward_compat.py
uv run python scripts/check_pua_consistency.py
uv run python scripts/check_dictionary_consistency.py
uv run python scripts/check_dictionary_versions.py
uv run python scripts/check_cli_flag_parity.py
uv run python scripts/check_phoneme_set_version.py
uv run python scripts/check_language_id_map_contract.py
uv run python scripts/check_inference_input_contract.py
uv run python scripts/check_short_text_contract.py
uv run python scripts/check_text_splitter_contract.py
uv run python scripts/check_streaming_api_contract.py
uv run python scripts/check_ssml_contract.py
uv run python scripts/check_speaker_encoder_contract.py
uv run python scripts/check_audio_format_contract.py
uv run python scripts/check_swift_g2p_contract.py
uv run python scripts/check_pt_dialect_contract.py

該当領域別の gate 早見表:

Python 変更箇所走らせるべき gate
chinese.py (loanword 関連)check_loanword_consistency.py + check_loanword_forward_compat.py
phonemize/ の PUA テーブルcheck_pua_consistency.py
言語コード追加・変更check_language_id_map_contract.py
辞書フォーマット変更check_dictionary_consistency.py + check_dictionary_versions.py
新規 CLI flag 追加check_cli_flag_parity.py
SSML 機能拡張check_ssml_contract.py
text_splitter.py 変更check_text_splitter_contract.py
streaming API 変更check_streaming_api_contract.py
speaker_encoder/ 変更check_speaker_encoder_contract.py
短テキスト戦略変更check_short_text_contract.py
PT BR/EU 切替変更check_pt_dialect_contract.py
Swift G2P 変更check_swift_g2p_contract.py
Show full SKILL.md (328 more words)Show less
7. Docker image の追随確認 (PR #443 の盲点)

src/python_run/piper_plus/ 変更時は、 ベースとなる docker image 2 個も touch されて いるかを確認 (Python ランタイムを wheel として再 install するだけで OK な場合と、 recipe を変える必要がある場合がある):

bash
echo "=== Docker image の touch 状況 ==="
for d in docker/python-inference docker/webui; do
  if git diff --name-only origin/dev...HEAD | grep -q "^${d}/"; then
    echo "  [TOUCHED  ] $d"
  else
    echo "  [UNTOUCHED] $d  ← Python wheel の bump で十分か確認"
  fi
done

# requirements.txt / pyproject 内の piper-plus pin が古いか確認
grep -rn 'piper-plus' docker/python-inference/ docker/webui/ 2>/dev/null | \
  grep -E '(==|>=|~=)' | head

確認すべき事項 (チェックリスト)

PR 提出前にメンタルチェック:

  • canonical Python の変更カテゴリ (A: 推論 / B: G2P / C: 学習) を分類した
  • 影響度 (Low / Medium / High) を判定した
  • Low 以外なら他 runtime の touch 状況を git diff --name-only で確認した
  • Medium なら Rust / Go / C# を同 PR で更新 (or follow-up issue 化)
  • High なら全 7 runtime + 2 docker を同 PR で更新 (or release blocker 立て)
  • 該当する scripts/check_*.py sync gate を local で pass
  • docs/spec/*.toml の version bump が必要か検討した
  • PR description に 「Python canonical + N 他ランタイム追随」 を明記
  • follow-up issue がある場合は PR description にリンク
  • ONNX I/O 変更なら /check-runtime-parity も併用した
  • 新規データファイル追加なら /check-new-runtime-asset も併用した

影響度別 quick action

Low (内部リファクタ・テスト追加・docstring)
text
PR description に以下を明記:
  > Python のみ変更。 他 runtime への追随不要 (内部リファクタ)。
  > /check-cross-runtime: skip (Low)
Medium (新機能・CLI flag・API field 追加)
  1. Rust + Go + C# を 同 PR で実装
  2. WASM / C++ / iOS / Kotlin / Swift G2P は follow-up issue 化
  3. PR description に follow-up issue link
bash
# follow-up issue を作る例
gh issue create --title "[follow-up] <feature> を WASM/C++/iOS/Kotlin/Swift に追随" \
  --body "PR #XXX で Python + Rust + Go + C# に実装済み。 残り 5 runtime に追随必要。"
High (config schema / data 形式 / 互換性破壊)
  1. 全 7 runtime + 2 docker を同 PR が原則
  2. 不可なら release blocker label を付けて release から外す
  3. docs/spec/*.toml の version bump
  4. docs/migration/ にマイグレーションガイド追加

トラブルシューティング

症状対処
Python 変更が大きすぎて何が canonical か分からないgit diff --stat origin/dev...HEAD で行数の多いファイルから 3 カテゴリ (A/B/C) に分類
影響度判定に迷う「他 runtime に同じバグが潜むか?」 を考える。 Yes なら Medium 以上
sync gate が無い領域を触ってしまったscripts/check_dictionary_consistency.py をテンプレートに新規 gate を作成 (/check-new-runtime-asset 参照)
全 runtime 同 PR が大きすぎるHigh なら release blocker、 Medium なら Rust/Go/C# だけ同 PR + follow-up issue
docker image の wheel pin が古いdocker/*/requirements.txt または pyproject.toml で piper-plus==X.Y.Z を bump
CI で sync gate fail該当 sync gate の PAIRS / mirror に追加する canonical entry が漏れている可能性

関連ドキュメント / skill

  • 姉妹 skill: /check-runtime-parity — ONNX グラフ I/O 限定
  • 姉妹 skill: /check-new-runtime-asset — 新規データファイル 7 箇所配布
  • 姉妹 skill: /check-loanword — ZH-EN loanword 7 mirror 同期
  • 姉妹 skill: /check-pua — PUA テーブル / fixture / config 整合
  • 仕様: docs/spec/inference-input-contract.toml
  • 仕様: docs/spec/short-text-contract.toml
  • 仕様: docs/spec/text-splitter-contract.toml
  • 仕様: docs/spec/phoneme-timing-contract.toml
  • マイグレーション: docs/migration/v1.11-to-v1.12.md
  • memory: feedback_data_asset_distribution.md (7 箇所配布の背景)
  • PR #391 / PR #443 — Python のみ修正 → N runtime 追随漏れ silent regression の参考事例
  • PR #384 — ZH-EN loanword schema 拡張で 5 runtime fail の事例

© ayutaz, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/check-cross-runtime of ayutaz/piper-plus.

Open the folder on GitHubat commit 421d066

Compare with similar skills

Check Cross Runtime next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Check Cross Runtime compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Check Cross Runtime this skillayutaz/piper-plus230—~3.5kAutomated safety check: PassMIT
Deploy To Tempsgotempsh/temps828—~1.3kAutomated safety check: NotesApache-2.0
Writing Dockerfilesancoleman/ai-design-components526—~3.2kAutomated safety check: NotesMIT
Build FlydslROCm/FlyDSL290—~1.3kAutomated safety check: NotesCustom licence
Sync Reviewthun-res/vlink116—~290Automated safety check: PassApache-2.0
Docker Scout AuditHabitat-Thinking/ai-literacy-superpowers114—~2kAutomated safety check: PassCustom licence

Similar skills

  • Deploy To Temps

    gotempsh/temps

    Deploy applications to the Temps platform with automatic framework detection, Dockerfile generation, and container orchestration.

    828 GitHub stars~1.3k tokensUpdated today
    DevOps & CloudAuto-check: notes
  • Writing Dockerfiles

    ancoleman/ai-design-components

    Writing optimized, secure, multi-stage Dockerfiles with language-specific patterns (Python, Node.js, Go, Rust), BuildKit features, and distroless images.

    526 GitHub stars~3.2k tokensUpdated 10 mo ago
    DevOps & CloudAuto-check: notes
  • Build Flydsl

    ROCm/FlyDSL

    Build and install FlyDSL (Flexible Layout Python DSL) on a remote host or Docker container.

    290 GitHub stars~1.3k tokensUpdated today
    DevOps & CloudAuto-check: notes
  • Sync Review

    thun-res/vlink

    审计 VLink 代码、doc、Doxygen、Python/C API、测试、示例、版本、 CMake、conanfile.py、Android.bp、vcpkg、packup 构建打包面、Agent 索引以及 .github 的 workflows、scripts、Docker、模板、Wiki 与治理 文件同步关系,定位 API、功能或工程入口变化后的遗漏和过时描述。公开…

    116 GitHub stars~290 tokensUpdated today
    DevOps & CloudAuto-check passed
  • Docker Scout Audit

    Habitat-Thinking/ai-literacy-superpowers

    A skill your agent uses when auditing Docker images in this project for CVEs, base image staleness, or remediation recommendations — covers all four TUI images (Go, Python, Kotlin, C)

    114 GitHub stars~2k tokensUpdated 18 days ago
    SecurityAuto-check passed
  • Flowfile Build and Environment Setup

    Edwardvaneechoud/Flowfile

    Recreates every Flowfile development and build environment from scratch, with exact version pins and an explanation of what each Makefile target really does.

    375 GitHub stars~7.3k tokensUpdated today
    DevelopmentAuto-check: notes

More from ayutaz/piper-plus

All 20 skills in this repo
  • Skill Health

    ayutaz/piper-plus

    .claude/skills//SKILL.md の frontmatter / referenced script 実在 / trigger 衝突 / .claude/hooks/.sh の executable+shebang を検査する meta-skill。

    230 GitHub stars~626 tokensUpdated yesterday
    Auto-check passed
  • Bump Deps

    ayutaz/piper-plus

    ORT / openjtalk / ruff のような cross-runtime に canonical sync が必要な依存関係を 1 コマンドで bump する read-mostly skill。

    230 GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed
  • Check Loanword

    ayutaz/piper-plus

    ZH-EN code-switching loanword の同期と forward-compat を 1 コマンドで検査。zhenloanword.json を編集したり 5 ランタイムのいずれかに新規エントリを追加する前後に呼ぶ。Python source を canonical とし、Rust×2 / Go / C / WASM / C++ の 6 mirror + Python…

    230 GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed
  • Check New Runtime Asset

    ayutaz/piper-plus

    新規データアセット (JSON/TOML) を追加した PR で「7 箇所の package metadata 更新が全て揃っているか」を 1 コマンドで確認。MANIFEST.in / pyproject package-data / Cargo features / npm files / C Content / Android assets / SPM resources の更新漏れを…

    230 GitHub stars~964 tokensUpdated yesterday
    Auto-check passed
  • Check PR Ready

    ayutaz/piper-plus

    PR 作成前の最終チェックリスト (lint/test/docs/CHANGELOG/未コミット確認)。/precheck の拡張版で、ドキュメント整合性も検証します。

    230 GitHub stars~687 tokensUpdated yesterday
    Auto-check passed
  • Check Pua

    ayutaz/piper-plus

    PUA テーブル / fixture / config の整合性を 1 コマンドで検査。pua.json を編集したり ɔɪ/œ̃/ɐ̃ のような multi-codepoint 音素を扱う前後に呼ぶ。docs/spec/pua-contract.toml の 4 不変条件を全部チェック。

    230 GitHub stars~768 tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Check Cross Runtime

What does Check Cross Runtime do?

Python canonical (src/pythonrun/piperplus/, src/python/pipertrain/, src/python/g2p/piperplusg2p/) を変更した PR で、 ONNX I/O 以外の追随漏れ (phonemizer / config schema / CLI flag / data 形式 / API 変更) を 7 ランタイム +…. Check Cross Runtime is an agent skill from ayutaz/piper-plus.

When should I use Check Cross Runtime?

Check Cross Runtime fits situations like: tasks that involve Containers.

How do I install Check Cross Runtime in Claude Code?

Run `npx skills add ayutaz/piper-plus --skill check-cross-runtime -a claude-code`. Or copy the skill folder (.claude/skills/check-cross-runtime in ayutaz/piper-plus) into .claude/skills/check-cross-runtime in your project. Claude Code loads it when a task matches its description.

How do I install Check Cross Runtime in Codex?

Run `npx skills add ayutaz/piper-plus --skill check-cross-runtime -a codex`. Or copy the skill folder (.claude/skills/check-cross-runtime in ayutaz/piper-plus) into .agents/skills/check-cross-runtime in your project. Codex loads it when a task matches its description.

Can I use Check Cross Runtime in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ayutaz/piper-plus --skill check-cross-runtime -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/check-cross-runtime, .gemini/skills/check-cross-runtime, .github/skills/check-cross-runtime and .opencode/skills/check-cross-runtime in your project.

What does Check Cross Runtime need to run?

Going by SKILL.md and its folder, Check Cross Runtime needs the command-line tools its instructions call (uv, git and gh). Our summary lists: Python 3; Docker. Its frontmatter pre-approves these tools: Bash(git diff *), Bash(git status *), Bash(grep *), Bash(ls *), Bash(python *), Bash(uv run *).

Does Check Cross Runtime access the network?

SKILL.md contains no URLs. Its commands use uv, git and gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Check Cross Runtime safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Check Cross Runtime use?

Check Cross Runtime is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Check Cross Runtime use?

About 3.5k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Check Cross Runtime?

Skills that share tags, products or a category with Check Cross Runtime: Deploy To Temps (gotempsh/temps, 828 stars), Writing Dockerfiles (ancoleman/ai-design-components, 526 stars), Build Flydsl (ROCm/FlyDSL, 290 stars) and Sync Review (thun-res/vlink, 116 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Check Cross Runtime?

ayutaz (a GitHub user) maintains it in ayutaz/piper-plus, which has 230 GitHub stars. The repository holds 20 skills in this directory. The repository was last updated on October 8, 2026.

Source: ayutaz/piper-plus on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.