docs(commands): 更新知识库分块命令中 --doc-id 的描述和注意事项

- 说明 --doc-id 在实际使用中为必需，避免服务器返回 HTTP 500 错误
- 明确指出应使用 doc list 命令中的文档级别 ID，拒绝使用 chunk list 中的每行 doc_id
- 新增说明向图片类型文档添加文本块会触发服务器错误，建议使用文本类型文档
- 对帮助文档中相关描述和备注进行了同步更新，增强使用指导性和准确性

Author: clark-fc

packages/commands/src/commands/knowledge/chunk-add.ts

@@ -23,7 +23,7 @@ const CHUNK_ADD_FLAGS = {
     type: "string",
     valueHint: "<id>",
     description:
-      "Owning document ID; required for table/image knowledge bases (the server rejects field-channel chunks without it), optional for document-type",
+      "Owning document ID from the doc list command; required in practice — the server returns HTTP 500 (dataId不能为空) without it even for unstructured KBs",
   },
   content: {
     type: "string",
@@ -74,7 +74,8 @@ export default defineCommand({
   flags: CHUNK_ADD_FLAGS,
   notes: [
     "Document / table / image knowledge bases are supported; audio-video ones are not.",
-    "Table/image knowledge bases require --doc-id — verified live: the server returns HTTP 500 (dataId不能为空) without it. Use the document-level id from the doc list command; the per-row doc_id in chunk list metadata is rejected (Index.InvalidParameter).",
+    "--doc-id is required in practice — verified live: even unstructured (document-type) KBs return HTTP 500 (dataId不能为空) without it. Use the document-level id from the doc list command; the per-row doc_id in chunk list metadata is rejected (Index.InvalidParameter).",
+    "Adding a text content chunk to an image-type document (jpg) triggers HTTP 500 (Index.SystemError) — the server does not support text chunks on image documents. Target a text-type document (docx/pdf/txt) instead.",
     "The API is idempotent but rate-limited to 10 calls per second — throttle batch scripts.",
     "The response carries no chunk id; list chunks afterwards to find the new one.",
     "For table/image knowledge bases use --field with Excel column headers as keys; values are passed through as strings.",

skills/bailian-cli/reference/knowledge.md

@@ -206,7 +206,7 @@ bl knowledge chat --message "Describe these images" --image https://example.com/
 | Flag                    | Type   | Required | Description                                                                                                                                               |
 | ----------------------- | ------ | -------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | `--index-id <id>`       | string | yes      | Knowledge base ID                                                                                                                                         |
-| `--doc-id <id>`         | string | no       | Owning document ID; required for table/image knowledge bases (the server rejects field-channel chunks without it), optional for document-type             |
+| `--doc-id <id>`         | string | no       | Owning document ID from the doc list command; required in practice — the server returns HTTP 500 (dataId不能为空) without it even for unstructured KBs    |
 | `--content <text>`      | string | no       | Chunk body text, up to 6000 chars (document-type); alternative to --content-file                                                                          |
 | `--content-file <path>` | string | no       | Read chunk body from a UTF-8 plain text file (.md/.txt etc.)                                                                                              |
 | `--title <text>`        | string | no       | Chunk title, up to 50 chars (document-type)                                                                                                               |
@@ -219,7 +219,8 @@ bl knowledge chat --message "Describe these images" --image https://example.com/
 #### Notes
 
 - Document / table / image knowledge bases are supported; audio-video ones are not.
-- Table/image knowledge bases require --doc-id — verified live: the server returns HTTP 500 (dataId不能为空) without it. Use the document-level id from the doc list command; the per-row doc_id in chunk list metadata is rejected (Index.InvalidParameter).
+- --doc-id is required in practice — verified live: even unstructured (document-type) KBs return HTTP 500 (dataId不能为空) without it. Use the document-level id from the doc list command; the per-row doc_id in chunk list metadata is rejected (Index.InvalidParameter).
+- Adding a text content chunk to an image-type document (jpg) triggers HTTP 500 (Index.SystemError) — the server does not support text chunks on image documents. Target a text-type document (docx/pdf/txt) instead.
 - The API is idempotent but rate-limited to 10 calls per second — throttle batch scripts.
 - The response carries no chunk id; list chunks afterwards to find the new one.
 - For table/image knowledge bases use --field with Excel column headers as keys; values are passed through as strings.