Replies: 3 comments 5 replies
-
|
I don't think it's necessary. If you tell an agent to read a memory after startup, it will do so. No need to force it |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
Yes but i have to tell it in a second prompt. It would be just a convenience factor i know ;) |
Beta Was this translation helpful? Give feedback.
3 replies
-
|
Update. Right now its working on 2 machines. I keep an eye on it the initial_prompt: "..." seems to work now, it is shown in the activate response as follow: |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Im not sure if we (claude and me) are missing something or if this would be a feature request. Im testing currently with Claude Desktop.
My problem is to tell Claude what memories to read.
Claude told me:
The core problem: The session_startup memory says it should be read first when SF2 activates, but Serena doesn't automatically trigger memory reads on project activation — the LLM only does this if it's explicitly reminded.So my solution was to create a startup memory which i activate with the prompt "startup". This works quite well.
If it is like that wouldnt it be a good idea if Serena would tell the LLM to read a special startup memory after project activation ?
P.S. using Sonnet 4.6
Beta Was this translation helpful? Give feedback.
All reactions