feat(feishu): support tables, table operations, image upload, and color text in doc tools #26222
Description
Problem
The feishu_doc tool only supported plain text and basic formatting. GFM tables, inline images, table manipulation, and text color were not supported.
Implemented (PR #26790)
Markdown rendering
- GFM tables now render as native Feishu tables via the Convert API (block types 31/32)
- Adaptive column widths based on cell content (CJK chars weighted 2x)
- Nested lists at any depth
- Documents >1000 blocks are automatically split into batches
New actions
| Action | Description |
|---|---|
upload_image |
Insert image from URL, data URI, local file path, or base64 string |
color_text |
Apply color/bold to a text block using [red]text[/red] markup |
insert_table_row / insert_table_column |
Insert row or column at given index |
delete_table_rows / delete_table_columns |
Delete rows or columns |
merge_table_cells |
Merge a rectangular cell range |
Image upload notes
- drive_route_token: When
parent_nodeis a document block ID, the API requires passingextra: { drive_route_token: docToken }. Per the API docs, certain upload scenarios require providing the token of the cloud document that the block belongs to. - Buffer must be passed directly to the upload form — using
Readable.from()causes Content-Length mismatch - File path detection uses
path.isAbsolute()instead of checking for/, since standard base64 alphabet includes/