repo-digest SKILL.md skillA
repo-digest from kendrick-stein/mcislab_deepread is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add kendrick-stein/mcislab_deepread/repo-digest@git:20260728.634c4b7Pins git:20260728.634c4b7 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- kendrick-stein/mcislab_deepread
- Path
skills/1-literature/repo-digest/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260728.634c4b7, committed Jul 28, 2026
- Size
- 8,064 bytes
- Repository stars
- 12
- License
- none declared
当已 digest 的论文带代码库（frontmatter `code` 非空 / 日志有 repo_candidate 标记）， 且属于系统/环境/基建类工作（贡献主要在实现里，如 CUA-Gym、WebHarbor、OpenRath、Crab）， 或 Supervisor 说"把代码库也读了""分析一下这个 repo"时：git clone 代码库、 并行派发多个 codex subagent 做静态分析，结合论文内容把实现细节写回 Papers/ 笔记。.
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
# Repo Digest — 论文代码库分析 ## Purpose 论文有 paper-digest，代码库没有对应物——但 CUA/AFE 方向的关键工作（环境 runtime、 benchmark、训练基建）贡献主要在系统实现里，论文正文往往只给 high-level 描述。 本 skill 把代码库 clone 下来做一次**静态分析**（不执行代码），将"论文怎么说 ↔ 代码怎么做" 的对照、论文没写的实现细节、复现路径写回对应 Papers/ 笔记，丰富报告细节、 为后续实验设计提供 grounding。 分析是重活，采用 **workflow + 多 codex subagent 并行**编排（复用 backfill 已验证的 `codex exec` 模式）：Codex 写各分析 pass，Claude 独立抽查核对后才入笔记 （Codex 写 / Claude 查，同 daily-papers codex digest 的分工纪律）。 ## Steps ### Step 1：定位 repo 与论文笔记 - 输入是 `Papers/` 笔记路径 → 读 frontmatter `code` 字段取 repo URL。 - 输入是 repo URL → Grep `Papers/` 找对应笔记；**无笔记则先跑 paper-digest**（本 skill 不替代论文消化，代码分析必须有论文笔记作对照基准）。 - 读取笔记的 Summary / Method / Key Results / Evidence Ledger，抽出 3-8 个 「要到代码里核对的点」：核心机制、关键超参、benchmark 生成/评测逻辑、 论文声称的系统能力。 ### Step 2：Clone 到 staging + 规模概览 ```bash mkdir -p Workbench/tmp/repo-digest && cd Workbench/tmp/repo-digest git clone --depth 1 <repo_url> <repo-name> cd <repo-name> && git rev-parse HEAD # 记录 commit sha find . -name "*.py" -o -name "*.ts" -o -name "*.js" | grep -v test | head -50 ``` - **>2GB 或含大量数据/权重文件**：停下向 Supervisor 确认（或用 sparse-checkout 只取代码目录）。 - 记录 commit sha——所有分析结论锚定到这个版本。 ### Step 3：生成分析任务清单（pass 划分） …
Read the whole file at its exact version.
How to install
mdr add kendrick-stein/mcislab_deepread/repo-digest@sha256:6e3fda39f82fd19bThe line at the top of this page pins the label git:20260728.634c4b7, 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_pqsouc2nijsnzxpn)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Jul 28, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8064 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
kendrick-stein/mcislab_deepread · 12 stars · license none declared · pushed 2026-10-19 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_pqsouc2nijsnzxpn GET https://markdownregistry.com/api/v1/resolve?ref=kendrick-stein/mcislab_deepread/repo-digest GET https://markdownregistry.com/api/v1/blob/6e3fda39f82fd19b00a944035d5742dab8fdb29e97612c103b41769b6822e9ea
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: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.