Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enhancement: Ability to select an agent as the default for new chats #5000

Open
1 task done
acehoss opened this issue Dec 15, 2024 · 7 comments
Open
1 task done

Enhancement: Ability to select an agent as the default for new chats #5000

acehoss opened this issue Dec 15, 2024 · 7 comments
Labels
enhancement New feature or request

Comments

@acehoss
Copy link

acehoss commented Dec 15, 2024

What features would you like to see added?

Maybe agents could appear in the preset list, allowing them to be selected as the default for new chats? Or some other way? But when starting a new chat, I'd like for an agent to be preselected.

More details

I have an agent made with the agent builder that I use for 95% of conversations, and every time I start a new chat I need to @ the agent in the chat box to select it.

Which components are impacted by your request?

No response

Pictures

No response

Code of Conduct

  • I agree to follow this project's Code of Conduct
@acehoss acehoss added the enhancement New feature or request label Dec 15, 2024
@dvejsada
Copy link

You can do this using a modelSpec https://www.librechat.ai/docs/configuration/librechat_yaml/object_structure/model_specs

You can set a specific agent as "endpoint" and even make it as default.

@acehoss
Copy link
Author

acehoss commented Dec 15, 2024

@dvejsada thanks for the suggestion! I gave it a try, but I couldn't figure out how to get it to select the agent I built in the agent builder. Additionally, all of the other endpoints disappeared. It seems like I'd have to manually set up model specs for everything I wanted to be able to use if I took this approach.

Additionally, my agent uses some custom actions, which don't appear to be configurable from the model specs.

I'd like to be able to select an agent that I built with the agent builder as the default.

@danny-avila
Copy link
Owner

danny-avila commented Dec 16, 2024

@acehoss I'm planning on implementing something similar to ChatGPT with the way they allow pinning GPTs:
firefox_ofKCnG9Sxx

Open to feedback, presets would also solve this.

@acehoss
Copy link
Author

acehoss commented Dec 16, 2024

@danny-avila That sounds even better than presets, the main thing I'd like is one-click activation of an agent. This would allow one-click activation of various agents, which is more powerful than default activation of one agent. 💯

@zeitderforschung
Copy link

zeitderforschung commented Dec 18, 2024

You can do this using a modelSpec https://www.librechat.ai/docs/configuration/libre
chat_yaml/object_structure/model_specs

You can set a specific agent as "endpoint" and even make it as default.

How did you manage to create an agent when modelSpecs were active? Changing the selection to "Agents" to show the Agent Builder was not possible once a modelSpec was configured. The menu entry "Agents" was there, but I could not select it. Without a modelSpec everything worked fine. I am already running the latest development build.

image

My environment:

ENDPOINTS=azureOpenAI,agents
EXPERIMENTAL_AGENTS=true

My librechat.yaml:

interface:
  endpointsMenu: true

endpoints:
  azureOpenAI:
  ...

modelSpecs:
  list:
    - name: "yourcompany-gpt"
      preset:
        default: true
        endpoint: "azureOpenAI"

@dvejsada
Copy link

You can do this using a modelSpec https://www.librechat.ai/docs/configuration/libre
chat_yaml/object_structure/model_specs

You can set a specific agent as "endpoint" and even make it as default.

How did you manage to create an agent when modelSpecs were active? Changing the selection to "Agents" to show the Agent Builder was not possible once a modelSpec was configured. The menu entry "Agents" was there, but I could not select it. Without a modelSpec everything worked fine. I am already running the latest development build.

image

My environment:

ENDPOINTS=azureOpenAI,agents
EXPERIMENTAL_AGENTS=true

My librechat.yaml:

interface:
  endpointsMenu: true

endpoints:
  azureOpenAI:
  ...

modelSpecs:
  list:
    - name: "yourcompany-gpt"
      preset:
        default: true
        endpoint: "azureOpenAI"

You can do modelSpec with endpoint "agents", which will allow you to create/edit agents. Once you have your agent created, just add the agent_id to keep the modelSpec endpoint for specific agent only (if preferred).

@Tsounguinzo
Copy link

You can do this using a modelSpec https://www.librechat.ai/docs/configuration/libre
chat_yaml/object_structure/model_specs
You can set a specific agent as "endpoint" and even make it as default.

How did you manage to create an agent when modelSpecs were active? Changing the selection to "Agents" to show the Agent Builder was not possible once a modelSpec was configured. The menu entry "Agents" was there, but I could not select it. Without a modelSpec everything worked fine. I am already running the latest development build.
image
My environment:

ENDPOINTS=azureOpenAI,agents
EXPERIMENTAL_AGENTS=true

My librechat.yaml:

interface:
  endpointsMenu: true

endpoints:
  azureOpenAI:
  ...

modelSpecs:
  list:
    - name: "yourcompany-gpt"
      preset:
        default: true
        endpoint: "azureOpenAI"

You can do modelSpec with endpoint "agents", which will allow you to create/edit agents. Once you have your agent created, just add the agent_id to keep the modelSpec endpoint for specific agent only (if preferred).

what do you mean by "add the agent_id", where do we get that id from and how do we add that Id in the model spec I cant find it in the docs also can I make that agent as default like we do with other models?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

5 participants