run-llama/llama_deploy
1
0
Fork 0
mirror of https://github.com/run-llama/llama_deploy.git synced 2026-08-24 21:01:26 -04:00
Code Issues 37 Packages Projects Releases 10 Wiki Activity

Custom Prompt for Orchestrator #27

New Issue
Closed
opened 2026-02-16 01:15:13 -05:00 by yindo · 4 comments
yindo commented 2026-02-16 01:15:13 -05:00
Owner
Copy Link
Copy Source

Originally created by @ryann-sportsbet on GitHub (Jun 30, 2024).

Currently, we only have two option for Orchestrator that either QueryPipeline or AgentOrchestrator without custom prompt parameters.

Can we have something similar like prompt in FunctionCallingAgentWorker?

The reason I'm asking because I have build a multi-agents with FunctionCallingAgentWorker and it is getting better response (both accuracy and meaningful ) compare to llama-agents.

Originally created by @ryann-sportsbet on GitHub (Jun 30, 2024). Currently, we only have two option for Orchestrator that either QueryPipeline or AgentOrchestrator without custom prompt parameters. Can we have something similar like prompt in FunctionCallingAgentWorker? The reason I'm asking because I have build a multi-agents with FunctionCallingAgentWorker and it is getting better response (both accuracy and meaningful ) compare to llama-agents.
yindo closed this issue 2026-02-16 01:15:13 -05:00
yindo commented 2026-02-16 01:15:15 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Jun 30, 2024):

You could likely migrate your existing multi agent system pretty easily by using the pipeline orchestrator, and then you have a distributed agent system 🙏

There are prompts for the agent orchestrator here, there are two main ones, plus the description of the "human", which is used for the final response
https://github.com/run-llama/llama-agents/blob/50a617b1b6545a662160916c757b3598a9ecc61b/llama_agents/orchestrators/agent.py#L31

@logan-markewich commented on GitHub (Jun 30, 2024): You could likely migrate your existing multi agent system pretty easily by using the pipeline orchestrator, and then you have a distributed agent system 🙏 There are prompts for the agent orchestrator here, there are two main ones, plus the description of the "human", which is used for the final response https://github.com/run-llama/llama-agents/blob/50a617b1b6545a662160916c757b3598a9ecc61b/llama_agents/orchestrators/agent.py#L31
yindo commented 2026-02-16 01:15:15 -05:00
Author
Owner
Copy Link
Copy Source

@ryann-sportsbet commented on GitHub (Jul 1, 2024):

@logan-markewich

Thank you sir.

I've implemented your suggestion by modifying both the summary and follow-up prompts with some adjustments to the human description. However, the results remain unchanged. In some instances, it appears that no agent is being called from the available agents, and the response seems to be based solely on the its knowledge

pipeline_orchestrator = PipelineOrchestrator(pipeline) control_plan = ControlPlaneServer(message_queue=message_queue, orchestrator=AgentOrchestrator( llm=llm, followup_prompt=test_followup_prompt(), summarize_prompt=test_multi_agent_summary_prompt(), human_description=test_human_description_prompt() ))

@ryann-sportsbet commented on GitHub (Jul 1, 2024): @logan-markewich Thank you sir. I've implemented your suggestion by modifying both the summary and follow-up prompts with some adjustments to the human description. However, the results remain unchanged. In some instances, it appears that no agent is being called from the available agents, and the response seems to be based solely on the its knowledge `pipeline_orchestrator = PipelineOrchestrator(pipeline) control_plan = ControlPlaneServer(message_queue=message_queue, orchestrator=AgentOrchestrator( llm=llm, followup_prompt=test_followup_prompt(), summarize_prompt=test_multi_agent_summary_prompt(), human_description=test_human_description_prompt() ))`
yindo commented 2026-02-16 01:15:15 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Jul 2, 2024):

You might need to give more complete code to properly debug. I can't do much with the above

@logan-markewich commented on GitHub (Jul 2, 2024): You might need to give more complete code to properly debug. I can't do much with the above
yindo commented 2026-02-16 01:15:15 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Sep 5, 2024):

We've done a major refactor, and I don't think this issue is relevant anymore

@logan-markewich commented on GitHub (Sep 5, 2024): We've done a major refactor, and I don't think this issue is relevant anymore
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
P0
P1
P2
bug
documentation
documentation
enhancement
example
good first issue
invalid
pull-request
Mirrored from GitHub Pull Request
 question
topic:CLI
topic:SDK
topic:SDK
topic:SDK
topic:apiserver
topic:control_plane
topic:message_queues
wontfix
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/llama_deploy#27
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?