BC web-to-FIM
将任意网页链接或本地文件一键转为结构化 Markdown，并三处存放到 Obsidian Vault、飞书云盘、腾讯 IMA 知识库。 支持的信源：(1) X/Twitter 推文、长文 Article、Thread 线程（逐字转录）；(2) 微信公众号文章（带图片，IMA 服务端抓取保留排版）； (3) 飞书 wiki 文档（v3.7.0 改用 lark-cli docs +fetch 抓取完整内容，WebFetch 截断严重已弃用）； (4) 小红书笔记；(5) 微博；(6) YouTube 视频； (7) 任意 HTML 网页（带图片，IMA 服务端抓取）；(8) 本地文件：PDF、Word、PPT、Excel、图片、音频等。 三处存放：Obsidian（本地 Markdown+frontmatter+tags）+ 飞书云盘（.md 原文件，用户身份上传立即可见）+ IMA 知识库（FIM知识库，AI 原生）。 v3.7.0 飞书 wiki 抓取改造：从 WebFetch（截断严重，200-1000 字符）改为 lark-cli docs +fetch（完整内容，3K-25K 字符），实测 33 篇文章中 19 篇被 WebFetch 截断。 v3.6.0 飞书存储改造：从应用身份创建在线 docx 改为 lark-cli drive +upload 上传 .md 原文件到用户云盘，用户在飞书"我的空间 > 云盘"立即可见。 IMA 智能路由：公众号/普通网页 → import_urls 保留图片；X/Twitter/飞书/GitHub → 纯文本笔记逐字转录。 飞书 wiki 原文链接优先转录：检查文章头部"原文链接"，有则优先抓取原文内容作为最终产物存三处，IMA 只存原文链接地址自动识别。 X 链接转录流程：飞书 wiki 中发现 X 原文链接时，用 x-tweet-fetcher 抓取完整长文（10K-15K 字符），失败时降级到飞书 wiki 内容。 批量模式支持断点恢复，Obsidian 写入失败自动 fallback。 工作流：自动识别 URL/文件类型 → 路由到最佳抓取工具 → 结构化 Markdown → 三处存放。 触发词：抓网页存飞书、web to feishu、url转文档、文件转飞书、存到ima、存到obsidian、web to fim、三处存放。 当用户明确要求将 URL 或本地文件转存为文档并存储到 Obsidian/飞书/IMA 时触发。 Do NOT use for 创建文档内容、编辑飞书文档正文、代码开发、非文档转存任务、普通文档格式转换（非三处存放意图）。
As a process C 52/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 15. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1122 chars, limit 1024 - warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - note
frontmatter-keyunknown frontmatter key "label" - note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "summary"
Process rating: all ten parameters 52/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 3 mutating operations with no state check
- 50When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (web, python) that frontmatter does not declare
- 100Steps. 44 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3298 tokens
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +3Description length 1121: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -4Absolute local paths (C:\Users, /home/…): not portable
- -212 emoji in the instructions: noise for the model
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 27 headings
- +3Step-by-step instructions: 44 items
- +4Has examples (12 code blocks)
- +4Reference files are cited in the instructions (1 of 3)
- +3All 5 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 54.