CLAUDE.md@docs/measurements/2026-10-30-dd-loop-triggerA
CLAUDE.md@docs/measurements/2026-10-30-dd-loop-trigger from kopp0510/claude-dd is a CLAUDE.md file: the standing instructions Claude Code reads in that repo.
mdr add kopp0510/claude-dd/docs/measurements/2026-10-30-dd-loop-trigger/CLAUDE.md@git:20261029.f2ab54aPins git:20261029.f2ab54a in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 14 of 14 checks passed
- Kind
- CLAUDE.md file
- Repository
- kopp0510/claude-dd
- Path
docs/measurements/2026-10-30-dd-loop-trigger/CLAUDE.md- Versions recorded
- 3
- Latest version
- git:20261029.f2ab54a, committed Oct 29, 2026
- First version recorded
- Oct 29, 2026
- Size
- 1,641 bytes
- Repository stars
- 12
- License
- MIT
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# docs/measurements/2026-10-30-dd-loop-trigger
`skills/dd-loop` 的觸發實測：全域 §3.9 改成明寫「呼叫 Skill `dd-loop`」之後，全新 session 會不會真的叫它（全域 CLAUDE.md 瘦身 S1-2）。
## 檔案
- `run.sh` —— 建拋棄式 repo（有一張 S1 `IN_PROGRESS` 的進度表）並跑 `claude -p`，四個正例各 2 次、一個反例 1 次。
旗標照根 CLAUDE.md「新增 Skill 步驟 5」，另加 `--settings '{"advisorModel":""}'`；只放行 `Skill Read Glob Grep`，只測叫不叫
- `check.sh` —— 從 jsonl 只抽判定欄位，欄位定義寫在它的檔頭。skill 本文不在 Skill 的 tool_result 裡（那裡只有 `Launching skill: dd-loop`），在下一則以 `Base directory for this skill:` 開頭的 user 訊息
- `results.tsv` —— `check.sh` 的輸出，被 `README.md` 與設計文件引用的證據；**不要在原處重跑 `claude -p` 蓋掉**，要重測就開新的日期目錄（S1 審查後用修正版 `check.sh` 對同一批 jsonl 重抽過一次，`README.md` 有記）
- `*.prompt.txt` —— 五個情境的 prompt 原文
- `README.md` —— 結果表、限制、context 量
## 此層慣例
- 原始 stream-json 不進 repo（init 訊息有編碼成 `-Users-<名字>-…` 的路徑、連接器清單），只留 `check.sh` 抽出來的欄位
- 跑之前要先 `./install-dd-pipeline.sh --force`：headless session 載入的是 `~/.claude/` 的部署版
## 與上層的關係
`docs/measurements/` 放實測原始數字；引用它的是 `docs/designs/2026-10-30-global-claude-md-slim-design.md` 的 S1 與 `DD_PIPELINE_ARCHITECTURE.md` 部署清單那句。Read the whole file at its exact version.
How to install
mdr add kopp0510/claude-dd/docs/measurements/2026-10-30-dd-loop-trigger/CLAUDE.md@sha256:3a4de39274760c8eThe line at the top of this page pins the label git:20261029.f2ab54a, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_2dwonkbb3ds2j27l)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, all committed on Oct 29, 2026.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261029.f2ab54a latest | 2026-10-29 | f2ab54a | 1,641 B | A | view · diff |
| git:20261029.6380120 | 2026-10-29 | 6380120 | 1,531 B | A | view · diff |
| git:20261029.f0e80ab | 2026-10-29 | f0e80ab | 1,428 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (1641 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
kopp0510/claude-dd · 12 stars · license MIT · pushed 2026-10-30 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_2dwonkbb3ds2j27l GET https://markdownregistry.com/api/v1/resolve?ref=kopp0510/claude-dd/docs/measurements/2026-10-30-dd-loop-trigger/CLAUDE.md GET https://markdownregistry.com/api/v1/blob/3a4de39274760c8e26cfa21cd271029ad0a3e1ee6672a041fe884b5c946bb8df
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: what a CLAUDE.md file is · keeping AGENTS.md and CLAUDE.md in sync · which agent file does what.