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

Use NER to make the general agent do stuff without specific prompting #559

Open
kongzii opened this issue Nov 22, 2024 · 0 comments
Open

Comments

@kongzii
Copy link
Contributor

kongzii commented Nov 22, 2024

Maybe NER (Named Entity Recognition) would be a good way to make the agent do stuff without over-prompting it.

Let's assume prompt like Get me balance of wallet address 0x2FC96c4e7818dBdc3D72A463F47f0E1CeEa0A2D0 with position id 38804060408381130621475891941405037249059836800475827360004002125093421139610 on Conditional Tokens Contract on Gnosis Chain..

Agent would first identify important keywords: balance of wallet address, position id , Conditional Tokens Contract , Gnosis Chain. This seems to work pretty good:

Screenshot by Dropbox Capture

Then use Tavily to google-up information about all of these:

First try with Tavily Research for prompt What is Conditional Tokens Contract Gnosis Chain?:

Screenshot by Dropbox Capture

If we would scrap also the references, then the agent would found out that:

Screenshot by Dropbox Capture

Then inject the information to the original prompt.

Then hope that the researched information + prompt is sufficient to complete the job. 😄

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

No branches or pull requests

1 participant