diff --git a/AGENT.md b/AGENT.md deleted file mode 100644 index 195df44..0000000 --- a/AGENT.md +++ /dev/null @@ -1,66 +0,0 @@ -# Agent Notes - -## Help Center Screenshot Processing - -Use the local Teable screenshot skill for docs screenshot work: - -```bash -python3 /Users/leo/.agents/skills/teable-doc-screenshots/scripts/stage_image.py \ - --input /tmp/clean.png \ - --output images/staged-field-preview/2026-05-11-topic.png \ - --padding-y 24 \ - --auto-fit-width -``` - -Final screenshot PNGs in the docs repo should be baked assets, not raw captures. - -- Keep raw screenshots and experiment outputs in `/tmp` or an ignored scratch directory. -- Commit only final PNGs that are referenced by Markdown or MDX. -- Use one standard canvas for normal docs images: `1600px` wide. -- Normal output uses `padding-x=48` and `padding-y=24`. -- `--auto-fit-width` may reduce horizontal padding to `24px` and upscale modestly when all of these are true: - - source width is at least `1200px` - - aspect ratio is between `1.35` and `2.6` - - ordinary rendered width would be below `1472px` - - required scale is at most `1.3` - - final height is at most `1000px` -- Very wide local UI strips should not be fit-width; leave them at normal size. -- Do not use legacy `wide`, `panel`, `small`, or `tall` presets for normal docs images. -- Dotted background, screenshot corner radius, and screenshot shadow are baked into the PNG. -- Docs CSS owns only the outer image border and small radius. - -In MDX, prefer: - -```mdx -Describe the UI state -``` - -Avoid random per-image widths such as `50%`, `65%`, or `100%`, and avoid -`docs-screenshot-full`, `docs-screenshot-panel`, `docs-screenshot-sm`, and -`docs-screenshot-tall`. - -Before committing screenshot batches, verify that every staged screenshot is -referenced and no experiment image is included: - -```bash -python3 - <<'PY' -import re, subprocess -from pathlib import Path - -refs = set() -for root in ("en", "zh"): - for path in Path(root).rglob("*"): - if path.suffix.lower() in {".md", ".mdx"}: - refs.update(r.lstrip("/") for r in re.findall(r"/images/staged-[^\s)\"']+\.png", path.read_text(errors="ignore"))) - -staged = subprocess.check_output(["git", "diff", "--cached", "--name-only"], text=True).splitlines() -images = {path for path in staged if path.startswith("images/staged-") and path.endswith(".png")} - -print("extra staged images", sorted(images - refs)) -print("missing staged image refs", sorted(refs - images)) -PY -``` diff --git a/en/basic/base.mdx b/en/basic/base.mdx index 2cbde55..c5d5059 100644 --- a/en/basic/base.mdx +++ b/en/basic/base.mdx @@ -39,11 +39,13 @@ Common operations include: ### Duplicate a Base to Another Space -Step 1: In the workspace, select the base you want to move; +Step 1: In the workspace, select the base you want to duplicate; Step 2: Click the menu icon and select the Copy Base option; -Step 3: A popup window will appear with a dropdown menu of available spaces where you can duplicate the base to. You need Creator permissions in the target space to move the base there. +Step 3: In the popup, choose the target space. You need Creator permissions in that space. + +For larger bases, Teable shows progress in the duplicate dialog while it copies the base structure, records, and attachments. Keep the dialog open until the copy finishes. Revision history and collaborators are not copied. + +Teable converts cross-base relation fields to single-line text fields in the exported file. + + ### Share a Base Bases can be shared through public links. The sharing scope can be the entire base or the currently selected item. diff --git a/zh/basic/base.mdx b/zh/basic/base.mdx index 0c93c54..b6c08c0 100644 --- a/zh/basic/base.mdx +++ b/zh/basic/base.mdx @@ -57,6 +57,8 @@ description: "创建、管理、导入导出数据库。通过 .tea 文件在 Te 第三步：在弹出的窗口中选择目标空间，需拥有目标空间的`可搭建`权限才能完成复制 +如果数据库较大，复制窗口会显示复制进度。请等复制完成后再关闭窗口。记录修改历史和协作者不会被复制。 + 复制数据库到另一个空间 + +跨数据库关联字段会在导出文件中转换为单行文本。 + + ### 分享数据库 数据库可以通过公开链接分享。分享范围可以是整个数据库，也可以是当前选中的某个节点。 diff --git a/zh/basic/space/im-integration.mdx b/zh/basic/space/im-integration.mdx index 432d472..3abd76f 100644 --- a/zh/basic/space/im-integration.mdx +++ b/zh/basic/space/im-integration.mdx @@ -68,7 +68,13 @@ im.message.receive_v1 测试通过后，将机器人添加到飞书群组，并在群内验证 CuppyClaw 是否能收到消息和响应。 -## 七、 常见排查 +## 七、 消息内容支持 + +CuppyClaw 可以处理飞书群组中的文字、图片、文件、音频、视频、表情和富文本消息。对于飞书卡片和分享类消息，它会提取其中可读的标题、正文和关键字段，作为 AI 对话上下文。 + +如果消息类型不在支持范围内，CuppyClaw 只能看到不支持的占位内容。遇到这种情况，可以改用文字说明，或把相关文件、图片直接给 CuppyClaw。 + +## 八、 常见排查 - **测试连接失败**：先确认 App ID 和 App Secret 是否来自同一个飞书应用。 - **群里无法使用机器人**：确认应用已发布，并且可用范围包含当前成员。