youngkingdom/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-16 01:48:40 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(workflow-generator): enhance the AI auto-creation flow end-to-end (#38175)

Browse Source 
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
This commit is contained in:
Crazywoola
2026-07-01 10:28:58 +08:00
committed by GitHub
parent 3ad06bebd9
commit 8809cc036d
85 changed files with 3386 additions and 500 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
api/openapi/markdown/console-openapi.md
View File

@ -9135,6 +9135,38 @@ Generate a Dify workflow graph from natural language
| 400 | Invalid request parameters | |
| 402 | Provider quota exceeded | |
### [POST] /workflow-generate/stream
Stream a Dify workflow graph (plan then result) via SSE
#### Request Body
| Required | Schema |
| -------- | ------ |
| Yes | **application/json**: [WorkflowGeneratePayload](#workflowgeneratepayload)<br> |
#### Responses
| Code | Description |
| ---- | ----------- |
| 200 | Server-Sent Events stream of plan/result events |
| 400 | Invalid request parameters |
### [POST] /workflow-generate/suggestions
Suggest example workflow-generator instructions for the tenant
#### Request Body
| Required | Schema |
| -------- | ------ |
| Yes | **application/json**: [WorkflowInstructionSuggestionsPayload](#workflowinstructionsuggestionspayload)<br> |
#### Responses
| Code | Description | Schema |
| ---- | ----------- | ------ |
| 200 | Suggestions generated successfully | **application/json**: [GeneratorResponse](#generatorresponse)<br> |
| 400 | Invalid request parameters | |
### [GET] /workflow/{workflow_run_id}/events
**Get workflow execution events stream after resume**
@ -21162,9 +21194,23 @@ can reuse its existing handler.
| current_graph | object | Existing draft graph to refine (cmd+k `/refine`); omit for create-from-scratch | No |
| ideal_output | string | Optional sample output for grounding | No |
| instruction | string | Natural-language workflow description | Yes |
| mode | string, <br>**Available values:** "advanced-chat", "workflow" | Target app mode for the generated graph<br>*Enum:* `"advanced-chat"`, `"workflow"` | Yes |
| mode | string, <br>**Available values:** "advanced-chat", "auto", "workflow" | Target app mode for the generated graph; 'auto' lets the backend classify the instruction<br>*Enum:* `"advanced-chat"`, `"auto"`, `"workflow"` | Yes |
| model_config | [ModelConfig](#modelconfig) | Model configuration | Yes |
#### WorkflowInstructionSuggestionsPayload
Payload for the workflow-generator instruction-suggestions endpoint.
Runs before the user picks a model, so the suggestions come from the
tenant's default model. The underlying generator never raises — an empty
``suggestions`` list is a valid 200 (soft-fail).
| Name | Type | Description | Required |
| ---- | ---- | ----------- | -------- |
| count | integer, <br>**Default:** 4 | Number of suggestions to return (1-6) | No |
| language | string | Optional language to write the suggestions in | No |
| mode | string, <br>**Available values:** "advanced-chat", "workflow" | Target app mode for the suggestions<br>*Enum:* `"advanced-chat"`, `"workflow"` | Yes |
#### WorkflowListQuery
| Name | Type | Description | Required |