You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
#### dialogflow:v3beta1
The following keys were added:
- schemas.GoogleCloudDialogflowCxV3beta1PlaybookInstruction.properties.guidelines.type (Total Keys: 1)
Copy file name to clipboardExpand all lines: docs/dyn/dialogflow_v3beta1.projects.locations.agents.playbooks.html
+12-6Lines changed: 12 additions & 6 deletions
Original file line number
Diff line number
Diff line change
@@ -129,7 +129,7 @@ <h3>Method Details</h3>
129
129
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
130
130
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
131
131
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
132
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
132
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
133
133
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
134
134
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
135
135
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -148,6 +148,7 @@ <h3>Method Details</h3>
148
148
},
149
149
],
150
150
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
151
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
151
152
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
152
153
{ # Message of single step execution.
153
154
"steps": [ # Sub-processing needed to execute the current step.
@@ -211,7 +212,7 @@ <h3>Method Details</h3>
211
212
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
212
213
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
213
214
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
214
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
215
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
215
216
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
216
217
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
217
218
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -230,6 +231,7 @@ <h3>Method Details</h3>
230
231
},
231
232
],
232
233
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
234
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
233
235
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
234
236
{ # Message of single step execution.
235
237
"steps": [ # Sub-processing needed to execute the current step.
@@ -361,7 +363,7 @@ <h3>Method Details</h3>
361
363
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
362
364
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
363
365
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
364
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
366
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
365
367
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
366
368
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
367
369
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -380,6 +382,7 @@ <h3>Method Details</h3>
380
382
},
381
383
],
382
384
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
385
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
383
386
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
384
387
{ # Message of single step execution.
385
388
"steps": [ # Sub-processing needed to execute the current step.
@@ -503,7 +506,7 @@ <h3>Method Details</h3>
503
506
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
504
507
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
505
508
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
506
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
509
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
507
510
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
508
511
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
509
512
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -522,6 +525,7 @@ <h3>Method Details</h3>
522
525
},
523
526
],
524
527
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
528
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
525
529
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
526
530
{ # Message of single step execution.
527
531
"steps": [ # Sub-processing needed to execute the current step.
@@ -603,7 +607,7 @@ <h3>Method Details</h3>
603
607
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
604
608
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
605
609
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
606
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
610
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
607
611
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
608
612
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
609
613
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -622,6 +626,7 @@ <h3>Method Details</h3>
622
626
},
623
627
],
624
628
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
629
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
625
630
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
626
631
{ # Message of single step execution.
627
632
"steps": [ # Sub-processing needed to execute the current step.
@@ -686,7 +691,7 @@ <h3>Method Details</h3>
686
691
{ # Playbook is the basic building block to instruct the LLM how to execute a certain task. A playbook consists of a goal to accomplish, an optional list of step by step instructions (the step instruction may refers to name of the custom or default plugin tools to use) to perform the task, a list of contextual input data to be passed in at the beginning of the invoked, and a list of output parameters to store the playbook result.
687
692
"createTime": "A String", # Output only. The timestamp of initial playbook creation.
688
693
"displayName": "A String", # Required. The human-readable name of the playbook, unique within an agent.
689
-
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish.
694
+
"goal": "A String", # Required. High level description of the goal the playbook intend to accomplish. A goal should be concise since it's visible to other playbooks that may reference this playbook.
690
695
"inputParameterDefinitions": [ # Optional. Defined structured input parameters for this playbook.
691
696
{ # Defines the properties of a parameter. Used to define parameters used in the agent and the input / output parameters for each fulfillment.
692
697
"description": "A String", # Human-readable description of the parameter. Limited to 300 characters.
@@ -705,6 +710,7 @@ <h3>Method Details</h3>
705
710
},
706
711
],
707
712
"instruction": { # Message of the Instruction of the playbook. # Instruction to accomplish target goal.
713
+
"guidelines": "A String", # General guidelines for the playbook. These are unstructured instructions that are not directly part of the goal, e.g. "Always be polite". It's valid for this text to be long and used instead of steps altogether.
708
714
"steps": [ # Ordered list of step by step execution instructions to accomplish target goal.
709
715
{ # Message of single step execution.
710
716
"steps": [ # Sub-processing needed to execute the current step.
0 commit comments