zvdy

regarding bedrock in aws, haven't checked it deeply, but i suspect that modifying the endpoint and touching a few routing/firewall rules should be enough to make it work just by...

> To my understanding, AWS Bedrock is not compatible with the OpenAI format, so it is necessary to rewrite the LLM Provider. yup, seems like anthropic sdk is necessary(and their...

> is this issue taken? I have a mock that works but got into a rabbit hole for other issues, so please take it As a note, last thing that...

@epicseven-cup friendly reminder on this, I have an issue that might be due to Claude's APIs/models and your PR might solve #322 I know you have started working on it...

#215 Solved this, give it a try

This is due to the `-x` flag, it redirects everything to `stderr` as you mentioned, In order to perserve the debugging information that `-x` prints, `2>&1` will just output it...

Just realized this was open, perhaps #215 solves what is mentioned here

I had a similar idea too, where kubectl-ai was installed singularly in a jumpbox and that jumbox would access _n_ clusters and troubleshoot on them. As this is a more...

@vitorguidi fixed the issue here, also worth noting the following: ```bash docker/base/start.sh 23:source /data/setup_clusterfuzz.sh 25:bash -ex /data/start_clusterfuzz.sh docker/chromium/base/start.sh 29:source /data/setup_clusterfuzz.sh 32:bash -ex /data/start_clusterfuzz.sh docker/chromium/builder/start.sh 25:source /data/setup_clusterfuzz.sh 27:bash -ex /data/start_clusterfuzz.sh docker/chromium/tests-syncer/start.sh...

@jonathanmetzman / @vitorguidi could you _gcburn_? https://github.com/google/clusterfuzz/pull/4318/checks?check_run_id=31549323008