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
{{ message }}
Repository navigation
Commit 6001c12
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: declarative-agents/workflow-samples/DeepResearch.yaml
+7-4Lines changed: 7 additions & 4 deletions
Original file line number
Diff line number
Diff line change
@@ -14,6 +14,9 @@
14
14
# - Knowledge Agent: Performs generic web searches.
15
15
# - Coder Agent: Able to write and execute code.
16
16
# - Weather Agent: Provides weather information.
17
+
#
18
+
# Example input:
19
+
# Find the current temperatures in Seattle and San Francisco, calculate the difference in Celsius and Fahrenheit, and recommend what clothing to pack for each city.
17
20
#
18
21
kind: Workflow
19
22
maxTurns: 500
@@ -264,14 +267,14 @@ trigger:
264
267
output:
265
268
messages: Local.Plan
266
269
input:
270
+
arguments:
271
+
team: =Local.TeamDescription
267
272
messages: |-
268
273
=UserMessage(
269
274
"Please briefly explain what went wrong on this last run (the root cause of the failure),
270
275
and then come up with a new plan that takes steps and/or includes hints to overcome prior challenges and especially avoids repeating the same mistakes.
271
-
As before, the new plan should be concise, be expressed in bullet-point form, and consider the following team composition
272
-
(do not involve any other outside people since we cannot contact anyone else):
273
-
274
-
{Local.TeamDescription}")
276
+
As before, the new plan should be concise, be expressed in bullet-point form, and only involve the team members already described
277
+
(do not involve any other outside people since we cannot contact anyone else).")
- WeatherAgent: Able to retrieve weather information
152
159
153
160
To make progress on the request, please answer the following questions, including necessary reasoning:
154
-
- Is the request fully satisfied? (True if complete, or False if the original request has yet to be SUCCESSFULLY and FULLY addressed)
161
+
- Is the request fully satisfied? (True if the requested work is complete and enough information is available to provide the final answer. A verified negative or empty result, such as confirming that a requested resource does not exist, can fully satisfy the request. False if work remains or if a negative or empty result may be caused by an execution, authorization, connectivity, or investigation failure.)
155
162
- Are we in a loop where we are repeating the same requests and / or getting the same responses from an agent multiple times? Loops can span multiple turns, and can include repeated actions like scrolling up or down more than a handful of times.
156
163
- Are we making forward progress? (True if just starting, or recent messages are adding value. False if recent messages show evidence of being stuck in a loop or if there is evidence of significant barriers to success such as the inability to read from a required file)
157
164
- Who should speak next? (select from: KnowledgeAgent, CoderAgent, WeatherAgent)
0 commit comments