@minded-ai/mindedjs 2.0.36-beta.5 → 2.0.36-beta.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/docs/low-code-editor/nodes.md +13 -13
- package/package.json +1 -1
|
@@ -67,20 +67,23 @@ agent.on(events.TRIGGER_EVENT, async ({ triggerName, triggerBody }) => {
|
|
|
67
67
|
|
|
68
68
|
## Prompt Nodes
|
|
69
69
|
|
|
70
|
-
Prompt nodes process input through LLM to generate intelligent responses or invoke tools.
|
|
70
|
+
Prompt nodes process input through LLM to generate intelligent responses or invoke tools.
|
|
71
71
|
|
|
72
|
-
|
|
72
|
+
### Context
|
|
73
|
+
- Prompt nodes automatically have access to previous nodes' output as context.
|
|
74
|
+
- You can also inject context from the memory object using the {memory.propertyName} placeholder.
|
|
73
75
|
|
|
74
|
-
|
|
76
|
+
### Properties
|
|
77
|
+
- `humanInTheLoop?: boolean` - When `true`, pauses execution after this node for human input before continuing. Use this property only if you need to gather an input from the user.
|
|
75
78
|
- `canStayOnNode?: boolean` - When `true`, allows the node to route back to itself for iterative processing. Usually combined with `humanInTheLoop: true` to allow for iteration over human input.
|
|
76
79
|
|
|
77
|
-
|
|
78
|
-
|
|
80
|
+
### Limitations
|
|
79
81
|
- Prompt nodes can not save information to memory directly. Rather the user & agent messages are stored in the messages array of the state object. Making user's input available to the next prompt nodes or tools in the form of input schema.
|
|
82
|
+
- Image recognition is only supported when a user attaches an image to the message. If you need to extract information from an image, use a tool with agent.llm to process the image. Return the result in the tool response so following prompt nodes can use it.
|
|
80
83
|
|
|
81
|
-
|
|
84
|
+
### Common Prompt Patterns
|
|
82
85
|
|
|
83
|
-
|
|
86
|
+
#### Extract Information from User
|
|
84
87
|
|
|
85
88
|
Use `humanInTheLoop: true` and `canStayOnNode: true` with `promptCondition` edges for iterative information gathering:
|
|
86
89
|
|
|
@@ -99,7 +102,7 @@ edges:
|
|
|
99
102
|
prompt: 'Has all required information been collected?'
|
|
100
103
|
```
|
|
101
104
|
|
|
102
|
-
|
|
105
|
+
#### Notify User
|
|
103
106
|
|
|
104
107
|
Use standard prompt nodes with `stepForward` edges for one-way notifications:
|
|
105
108
|
|
|
@@ -115,11 +118,8 @@ edges:
|
|
|
115
118
|
type: stepForward
|
|
116
119
|
```
|
|
117
120
|
|
|
118
|
-
### Configuration
|
|
119
|
-
|
|
120
|
-
**Placeholders:** Use `{memory.propertyName}`, `{memory.customer.name}`, or `{system.currentTime}` in prompts.
|
|
121
|
-
|
|
122
|
-
**LLM Configuration:**
|
|
121
|
+
### Custom LLM Configuration
|
|
122
|
+
Provide a custom LLM configuration to the prompt node using the `llmConfig` property.
|
|
123
123
|
|
|
124
124
|
```yaml
|
|
125
125
|
llmConfig:
|