context length exceeded error when trying to browse linkedin.

[sangee2004]

Steps to reproduce the problem:

1. Execte the following script to print details of the top 3 job picks from linkedin - gptscript --workspace ~/myworkspace --disable-cache test_browse_linkedin.gpt

Tools: github.com/gptscript-ai/browser, sys.workspace.read

Go to www.linkedin.com/jobs
Get the page link for Top job picks
For each of the page links got from previous step, get the details of the job.

I get the following error :

12:18:27 sent     [main]
         content  [1] content | Waiting for model response...2024/05/09 12:18:29 error, status code: 400, message: This model's maximum context length is 128000 tokens. However, your messages resulted in 214214 tokens (213645 in the messages, 569 in the functions). Please reduce the length of the messages or functions.
Closing the server

Debug logs:

12:16:48 started  [main]
12:16:48 sent     [main]
         content  [1] content | Waiting for model response...
         content  [1] content | tool call browse -> {"website":"https://www.linkedin.com/jobs"}
12:16:50 started  [browse(2)] [input={"website":"https://www.linkedin.com/jobs"}]
12:16:50 launched [service][https://raw.githubusercontent.com/gptscript-ai/browser/5da6f62858dfb5850646aeaf9aa65dff757eec4c/tool.gpt:91] port [10803] [gptscript sys.daemon /usr/bin/env npm --prefix /Users/sangeethahariharan/Library/Caches/gptscript/repos/5da6f62858dfb5850646aeaf9aa65dff757eec4c/node21 run server]
2024/05/09 12:16:50 error in request to [http://127.0.0.1:10803/browse] [404]: 404 Not Found
sangeethahariharan@Sangeethas-MBP gptscript % 
> gptscript-browser-plugin@1.0.0 server
> ts-node src/server.ts

Server is listening on port 10803
Closing the server

sangeethahariharan@Sangeethas-MBP gptscript % gptscript --workspace ~/myworkspace --disable-cache test_browse_linkedin.gpt 
12:17:50 started  [main]
12:17:50 sent     [main]
         content  [1] content | Waiting for model response...
         content  [1] content | tool call browse -> {"website":"https://www.linkedin.com/jobs"}
12:17:52 started  [browse(2)] [input={"website":"https://www.linkedin.com/jobs"}]
12:17:52 launched [service][https://raw.githubusercontent.com/gptscript-ai/browser/5da6f62858dfb5850646aeaf9aa65dff757eec4c/tool.gpt:91] port [11186] [gptscript sys.daemon /usr/bin/env npm --prefix /Users/sangeethahariharan/Library/Caches/gptscript/repos/5da6f62858dfb5850646aeaf9aa65dff757eec4c/node21 run server]

> gptscript-browser-plugin@1.0.0 server
> ts-node src/server.ts

Server is listening on port 11186
{ website: 'https://www.linkedin.com/jobs' }
12:18:00 ended    [browse(2)]
12:18:00 continue [main]
12:18:00 sent     [main]
         content  [1] content | Waiting for model response...
         content  [1] content | tool call getpagecontents -> {}
12:18:01 started  [getPageContents(3)] [input={}]
{}
12:18:06 ended    [getPageContents(3)] [output=0 notifications total
---------------------

content...

12:18:06 continue [main]
12:18:07 sent     [main]
         content  [1] content | Waiting for model response...
         content  [1] content | tool call getpagecontents -> {}
12:18:22 started  [getPageContents(4)] [input={}]
{}
12:18:27 ended    [getPageContents(4)] [output=0 notifications total
---------------------

content...

12:18:27 continue [main]
12:18:27 sent     [main]
         content  [1] content | Waiting for model response...2024/05/09 12:18:29 error, status code: 400, message: This model's maximum context length is 128000 tokens. However, your messages resulted in 214214 tokens (213645 in the messages, 569 in the functions). Please reduce the length of the messages or functions.
Closing the server