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
Idea: We could create a configuration model like this
# one stdio-based MCP server
quarkus.langchain4j.mcp.server.server1.transport-type=stdio
quarkus.langchain4j.mcp.server.server2.command=npm exec @modelcontextprotocol/[email protected]
# and one HTTP-based MCP server
quarkus.langchain4j.mcp.server.server2.transport-type=http
quarkus.langchain4j.mcp.server.server2.sse-url=http://localhost:3000/sse
and Quarkus would automatically generate a Tool provider that is backed by these two servers and register it in CDI to be picked up by AI services automatically
The text was updated successfully, but these errors were encountered:
This can be done after we get a LC4j version with langchain4j/langchain4j#2240 in
Idea: We could create a configuration model like this
and Quarkus would automatically generate a Tool provider that is backed by these two servers and register it in CDI to be picked up by AI services automatically
The text was updated successfully, but these errors were encountered: