Skip to content

LangGraph Toolbox local invoke fails due to missing model deployment #757

@Menghua1

Description

@Menghua1

Description:
When testing Scenario 3: LangGraph Toolbox (Responses), after azd ai agent run successfully started locally, executing the local invocation command azd ai agent invoke --local --protocol responses "What tools do you have?" failed. The local log showed the error: "DeploymentNotFound. The API deployment for this resource does not exist". See the image below for details.
Image
Image

Steps to Reproduce:

  1. mkdir langgraph-toolbox and cd langgraph-toolbox
  2. Run azd ai agent init --manifest "https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox/agent.manifest.yaml" and cd toolbox-langgraph.
  3. Run azd up.
  4. Run $env:TOOLBOX_ENDPOINT = (azd env get-value TOOLBOX_AGENT_TOOLS_MCP_ENDPOINT).
  5. Run azd ai agent run.
  6. From a second terminal, run azd ai agent invoke --local --protocol responses "What tools do you have?".

Environment:

Expected Behavior:
Local invoke can succeed.

@JeffreyCA and @RickWinter for notification.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions