Live survey304 models · 14 providers · 23,592 auditsAudited implement · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

feishu-doc-scraper

skill · daymade/claude-code-skillsMIT licenceAudited 13 Nov 2026 · source c5deba6fddc9

★ 1,530pushed today
!CAUTIONGrade B · 85/100 · 8 findings · c5deba6
1,530repo stars · whole daymade/claude-code-skills

Is it safe? With care — CAUTION, grade B (85/100). We read this skill's source at commit c5deba6fddc9 and observed network access and a hard-coded secret (see the findings). The audit raised 8 findings, 0 critical or high. What it does: Extracts Feishu/Lark Docs, Wiki, Sheets, and Minutes (妙记) transcripts into faithful local Markdown via the lark-cli API — no LLM paraphrasing, browser-DOM fallback when lark-cli can't reach content. Use whenever the…

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “feishu-doc-scraper (daymade/claude-code-skills)”. Retrieved 13 Nov 2026. https://modelsatlas.com/skills/daymade/feishu-doc-scraper

Trust audit

Audit 0.4.1 · source read 13 Nov 2026

Trust score85/100!CAUTION B
0 critical0 high6 medium2 low0 info

CAUTION means the audit found things worth knowing before you trust this skill. Grade B is shared by 17,406 of 18,389 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkDeclared5 observations
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsFoundSee the findings; values are never shown

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS1
L1 Static sourceSecrets, path traversal, supply chain, hidden codeWARN7
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (8)

MEDIUMObfuscation / stealthobf.zero_width · scripts/feishu_dom_capture.js:53
F-07 Obfuscation / stealthCWE-506 · CWE-94Layer L1
return result.replace(/[]/g, '');
Why
No further detail recorded for this rule.
Fix
—
MEDIUMObfuscation / stealthobf.zero_width · scripts/feishu_dom_capture.js:88
F-07 Obfuscation / stealthCWE-506 · CWE-94Layer L1
.map(c => (c.innerText || '').replace(/[\n]/g, ' ').trim());
Why
No further detail recorded for this rule.
Fix
—
MEDIUMObfuscation / stealthobf.zero_width · scripts/feishu_dom_capture.js:110
F-07 Obfuscation / stealthCWE-506 · CWE-94Layer L1
const text = (block.innerText || '').replace(/[]/g, '').trim();
Why
No further detail recorded for this rule.
Fix
—
MEDIUMObfuscation / stealthobf.zero_width · scripts/restore_docx_headings.py:54
F-07 Obfuscation / stealthCWE-506 · CWE-94Layer L1
_ZERO_WIDTH = ""
Why
No further detail recorded for this rule.
Fix
—
MEDIUMHard-coded secretssecret.generic · tests/fixtures/source-cards.html:1
F-03 Hard-coded secretsCWE-798 · CWE-321Layer L1
<title id="…">Attachment toolkit</title><h1 id="…">Section 01</h1><figure id="…" view-type="…"><source id="Synt…
Why
No further detail recorded for this rule.
Fix
—
MEDIUMHard-coded secretssecret.generic · tests/test_attachment_archives.py:20
F-03 Hard-coded secretsCWE-798 · CWE-321Layer L1
TOKEN = '…'
Why
No further detail recorded for this rule.
Fix
—
LOWNetwork egressnet.raw_ip · references/lark-cli-api-extraction.md:25
F-04 Network egressCWE-200 · CWE-319Layer L1
[lark-cli] [WARN] proxy detected: https_proxy=http://127.0.0.1:1082 — requests
Why
No further detail recorded for this rule.
Fix
—

Audit 0.4.1 · source sha c5deba6fddc9 · 13 Nov 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherdaymade GitHub userOwner of the repository this skill is read from
  • Source folderdaymade/claude-code-skills/feishu-doc-scraperThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdictc5deba6fddc9A new audit runs when the source changes
  • Repository statusNot archivedPushed today
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Extracts Feishu/Lark Docs, Wiki, Sheets, and Minutes (妙记) transcripts into faithful local Markdown via the lark-cli API — no LLM paraphrasing, browser-DOM fallback when lark-cli can't reach content. Use whenever the…

---
name: feishu-doc-scraper
description: >-
  Extracts Feishu/Lark Docs, Wiki, Sheets, and Minutes (妙记) transcripts into faithful local Markdown
  via the lark-cli API — no LLM paraphrasing, browser-DOM fallback when lark-cli can't reach
  content. Use whenever the source is a Feishu/Lark URL and fidelity matters — 导出飞书文档/合集/妙记转写, 把飞书
  wiki 转 markdown, or archiving a collection — even if the user just says "save this". Also covers
  owner-exported .docx → Markdown.
compatibility: Primary path needs the `lark-cli` binary (npm `@larksuite/cli`; verified 1.0.32, 2026-05, and re-verified 1.0.80, 2026-08 — the `.data.markdown` field is null on 1.0.80 and the pandoc/`source.html` path in step 3 is load-bearing there) authenticated to the target tenant. Fallback path needs a browser automation surface with an authenticated session (Chrome DevTools MCP / Browser Use / Computer Use). docx path needs `python-docx` and a docx→md converter (the bundled doc-to-markdown skill or pandoc).
argument-hint: "[feishu-url-or-output-path]"
---

# Feishu Doc Scraper

Extract a Feishu/Lark source into faithful local Markdown. **Prefer the lark-cli API** — it extracts the body programmatically (no model paraphrasing), follows a collection's reference graph, and reads permission boundaries from error codes instead of guessing. Treat the rendered browser page as a *fallback*, not the source of truth: in real collection-scraping work the API path consistently does the whole job while the browser path is never needed.

## Scope (read this first)

This skill's contract is **faithful per-source Markdown + a record of what was extracted**. Naming, indexing, deduplication and knowledge ownership belong to the recipient's archive workflow. For a filing request, including permission to choose a suitable location, first follow [archive destination handoff](references/archive-destination-handoff.md). Temporary extraction may proceed while the destination is unresolved; permanent filing and uploads follow the host's selected ownership and storage contract.

**For every document read or archived, load [comments and feedback](references/comments-and-feedback.md).**
It owns the body-plus-discussion workflow, selected scope, body-only exception,
snapshot commands, interpretation and coverage checks.

**Extraction and durable storage are separate decisions.** A downloaded MP4/XLSX/DOCX/image is a working copy, not evidence that the file belongs in Git or Git LFS. For a knowledge-base archive, default to:

- Git: faithful Markdown plus structured CSV/JSON/HTML, source locators, revision/permission state, byte counts, MIME types, and hashes.
- Platform original: Feishu remains the source of record for raw binary attachments when the stable document/file token is still retrievable. A local download is an optional cache and may be absent on a fresh clone.
- NAS retention: when the selected archive already uses an authorized NAS backend, follow the [NAS storage contract](references/archive-storage-contract.md#nas-retained-external-media), including original source identity and full retained-file integrity.
- OSS/object storage: only when the archive needs an independent durable copy or the platform source is not a reliable long-term retrieval path. Uploading is a separate authorized operation, not a fallback the extractor chooses by itself.

Before filing an archive, declare this split in an artifact manifest and run the bundled storage validator. The complete schema and examples are in **[references/archive-storage-contract.md](references/archive-storage-contract.md)**.

Keep source access instructions with the authorized destination's material entry,
following the destination handoff; return them with a password-protected source
link. Generate aggregate statistics on demand under the storage contract.

Select the delivery scope from the user's requested result before loading fallback
or reader procedures. For original-file downloads, use
[attachment archives](references/attachment-archives.md): fetch the body first,
enumerate its actual resource types, then load only those handlers. Record body,
attachments, independent retention and reader acceptance separately. An attachment
batch alone never establishes completion of the document's reference graph.

For website links, Bilibili, YouTube or direct-media links in the body, load
[external video discovery and handoff](references/external-video-resources.md).
Enumerate clickable and plain-text URLs on every captured body; opaque websites
remain link locators until selected for inspection. Use the reference's batch
scope/volume preflight before mass acquisition or cloud creation. Discovery,
text capture and media acquisition have separate completion states. Delegate
authorized acquisition to the installed platform/storage owner with the user's
selected part or playlist range.

For failed media exports, original-preview retrieval, or an Obsidian reading handoff, load
[original media and reader handoff](references/original-media-and-reader.md).
Use its verified-preview downloader and API-capture renderer before expanding
the batch. Its Delivery completion gate (`check_reader_delivery.py finalize`)
is the reader-delivery completion entry; conversion, byte and search success do
not replace a current actual-reader evidence record.
For a user-selected personal favorites destination, its source catalog producer
hands off to favorites-search without changing this Skill's filing ownership.

## Choose the path

Before the first linked-resource read, load `lark-cli-router` and follow its
[source account binding](https://github.com/daymade/claude-code-skills/blob/main/lark-cli-router/references/source-account-binding.md).
Retain the selected profile and identity for every branch below.

```
What is the source?
├── Minutes / 妙记 URL, or a document references one → PATH C: Minutes transcript
├── Owner-exported .docx → PATH B: faithful Markdown
└── Other Feishu/Lark URL (wiki / docx / sheets / base)

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit c5deba6

Download folder ↗
uv run --script <skill-dir>/scripts/download_original_preview.py \
uv run --script <skill-dir>/scripts/render_api_capture.py \
uv run --script <skill-dir>/scripts/check_reader_delivery.py inspect \
uv run --script <skill-dir>/scripts/check_reader_delivery.py record-reader \
uv run --script <skill-dir>/scripts/check_reader_delivery.py finalize \
uv run --script <skill-dir>/scripts/build_manual_catalog.py <source-archive-root> \

Commands as the repository documents them: shown, never run by us. Our verdict covers commit c5deba6fddc9 only.

Works in which agents?

What its own text says, not a test

AgentStatusMore
Claude CodeMentioned in its text, not testedSkills for Claude Code · About Claude Code
“Mentioned” means the skill’s text names the agent. We have not run it there.

Audit history

Every audit, every source change

  1. !CAUTION B 85 / 100

    first audit · source c5deba6fddc9

README badge

For maintainers · shows the latest audit

ModelsAtlas: CAUTION BMODELSATLASCAUTION · B
[![ModelsAtlas: CAUTION B](https://modelsatlas.com/skills/daymade/feishu-doc-scraper/badge.svg)](https://modelsatlas.com/skills/daymade/feishu-doc-scraper/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Add this badge to your README

[![ModelsAtlas: CAUTION B](https://modelsatlas.com/skills/daymade/feishu-doc-scraper/badge.svg)](https://modelsatlas.com/skills/daymade/feishu-doc-scraper/)

Questions about feishu-doc-scraper

What does the feishu-doc-scraper skill do?
Professional Claude Code skills marketplace featuring production-ready skills for enhanced development workflows.
Is feishu-doc-scraper safe to install?
With care. The audit graded it B (85/100) and found 8 things worth knowing before you trust this skill, listed below with the exact line each was found on.
What can feishu-doc-scraper access on my machine?
The audit observed that it reaches the network. Each of those is consistent with what it says it does. Secrets in the source: found — see the findings.
What do I need installed to use feishu-doc-scraper?
Its own instructions reference lark-cli and python-docx. Dependencies are pinned to exact versions.
Which assistants does feishu-doc-scraper work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (c5deba6fddc9), read on 2026-11-13. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.