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.
npx agentmods add instructions/ayutaz/piper-plus/claude-mdgit clone --depth 1 https://github.com/ayutaz/piper-plusWrote 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.
[](https://agentmods.dev/instructions/ayutaz/piper-plus/claude-md)<a href="https://agentmods.dev/instructions/ayutaz/piper-plus/claude-md"><img src="https://agentmods.dev/badge/instructions/ayutaz/piper-plus/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.14767 | $0.14767 |
| Opus 5 | $0.07383 | $0.07383 |
| Sonnet 5 | $0.02953 | $0.02953 |
| Haiku 4.5 | $0.01477 | $0.01477 |
Grade B, and why
piper-plus CLAUDE.md scanned grade B with 1 finding 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 5d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **C++**: `cmake -B build -DBUILD_TESTS=ON && cmake --build build -j && ctest --test-dir build` (ORT/OpenJTalk は cmake が自動 DL、sudo 不要)。 How it starts
The opening of the file, as written. The whole thing — 564 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Piper TTS - プロジェクト概要
VITS ベースの高品質ニューラル TTS。6言語マルチリンガルモデルを学習済み、8言語の G2P コードを全7ランタイム (Python/C#/Rust/Go/JS-WASM/C++/CLI) に実装済み。
ブランチ: dev (v1.12.0 リリース済み)。v1.12.0 Breaking changes: docs/migration/v1.11-to-v1.12.md。
📦 アーカイブ: zero-shot v7 学習成果 (feat/zero-shot-tts)
⚠️ 現在の作業ブランチは
dev(上の「ブランチ」欄参照)。このセクションは過去の zero-shot 学習成果と引き継ぎ用情報のアーカイブです。新規作業をfeat/zero-shot-tts上で行う必要はありません。
dev の最新 (5fceade、Release v1.12.0) を取り込み、Zero-Shot TTS の累積差分を 1 コミットに集約してリベース完了。origin/feat/zero-shot-tts に push 済み。
その後、multi-6lang zero-shot スクラッチ学習 (v7) を完遂中。リベース時に廃止された
Multi-scale FiLM を MBiSTFTGenerator に復活させ、DDP-synced NaN skip / speaker_embedding
L2 再正規化 / dino_center 汚染防御 を実装。詳細は
docs/design/multi-6lang-zero-shot-v7-training-results.md 参照。
🚚 環境を引き継ぐ場合:
docs/handoff/zero-shot-tts-handoff-2026-06-20.mdに ゼロからの環境構築・データセット復元・学習再開・FT・評価の全手順をまとめています。 モデル本体は HF (private) で配布:ayousanz/piper-plus-zero-shot-multi-6lang-v7/ayousanz/piper-plus-zero-shot-tsukuyomi/ayousanz/campplus-onnx(CAM++ mirror、引き継ぎ専用)
完了済みアセット (v7 multi-6lang スクラッチ + Tsukuyomi FT、2026-05-21 時点)
v7 学習は epoch 32 で停止。続けて Tsukuyomi zero-shot FT 500 epoch 完走 (2026-05-21 00:07〜00:58)。再開コマンドは下方「新 GPU での学習再開」参照。
📦 HF リポジトリ (private) — 2026-06-20 アップロード:
ayousanz/piper-plus-zero-shot-multi-6lang-v7— v7 resume ckpt + singlespk ckpt + ONNX (ep32) + 評価再現セット (compute_secs.py + 参照 emb + 合成 wav) + TensorBoard events + 過去 epoch ONNX (ep10/ep20)ayousanz/piper-plus-zero-shot-tsukuyomi— Tsukuyomi FT ckpt + ONNX + 合成サンプル + 評価再現セットayousanz/campplus-onnx— CAM++ Speaker Encoder (Apache-2.0、ModelScope iic ミラー、引き継ぎの簡便化用)
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.
- 5d ago First seen · 564 lines · 14,767 tokens per session scan B 54d6eec04382
piper-plus CLAUDE.md is an instructions file published in the GitHub repository ayutaz/piper-plus (204 stars, last pushed 3d ago), licensed MIT. It adds 14,767 tokens to every session, about $0.0738 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
DeepSpeed AGENTS.md
AGENTS.md instructions for deepspeedai/DeepSpeed, covering deepspeed project rules, commit & ci requirements, code change discipline, test discipline and tool caveats.
huggingface_hub AGENTS.md
AGENTS.md instructions for huggingface/huggingface_hub, covering agent guide for huggingfacehub, project overview, setup, key commands and code structure.
PINA AGENTS.md
AGENTS.md instructions for PINA-org/PINA, covering pina — physics-informed neural architectures, quick reference, workflow: problem → model → solver → trainer, problem types and condition types.
DeepSpeed CLAUDE.md
Claude Code instructions for deepspeedai/DeepSpeed, covering deepspeed project rules, commit & ci requirements, code change discipline, test discipline and tool caveats.
diffusers AGENTS.md
AGENTS.md instructions for huggingface/diffusers, a project described as: 🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
diffusers CLAUDE.md
Claude Code instructions for huggingface/diffusers, a project described as: 🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.