I've been using opencode go as my LLM provider and...
# generalc
clean-postman-32056
09/10/2026, 4:56 AMI've been using opencode go as my LLM provider and tried to pick up where I left off several days ago and found this new issue waiting for me. Aparently opencode go now requires special headers. Is there a way for me to configure this in the advanced section of the LLM config? github.com/aaif-goose/goose/issues/11900 and github.com/vastsa/PI-Desktop/issues/48
w
wooden-umbrella-84098
09/10/2026, 1:30 PMwooden-umbrella-84098
09/10/2026, 1:30 PMAs a tmp. workaround (working, confirmed) add this to Extra Headers in LLM settings
{"x-opencode-session":"facade00-0000-4000-a000-000000000000"}
c
clean-postman-32056
09/10/2026, 4:01 PMSo, we still don't have the ability to hot load changes on preexisting conversations?
clean-postman-32056
09/10/2026, 4:01 PMMaking those changes does nothing unless it's a new conversation
clean-postman-32056
09/10/2026, 4:02 PMAnd, thanks Robert for your time. I appreciate you.
w
wooden-umbrella-84098
09/10/2026, 4:05 PMhmm...if you create a new LLM profile can you switch to it mid-session? I think that should work
c
clean-postman-32056
09/10/2026, 4:08 PMNo, it doesn't. Entering those extra headers works on new conversations but my preexisting one still throws the same error. It's most likely because the conversation is cached, I learned that on another bug a few months ago. It's really something that needs to be addressed.
w
wooden-umbrella-84098
09/10/2026, 4:12 PMhmm makes sense. Is there an issue open?
c
clean-postman-32056
09/10/2026, 4:13 PMI have no idea honestly
c
calm-artist-7206
09/10/2026, 5:57 PMNeal, you can tell your agent to go find the logs of the old conversation and understand it, and continue from there
But I agree, we fixed the most problematic “frozen in the past “ behavior, but i don’t know offhand about extra headers
2 Views