Repository Issue Activity (beta)

#801 gpt-5.6 models: every request 400s when functions are enabled (function tools rejected on /v1/chat/completions)

#799 docs: hosted API_BASE_URL example (not Ollama / not LiteLLM)

#793 Function-calling feature executes LLM-proposed shell commands with no confirmation, unlike the tool's own --shell mode

#578 lmstudio ?

#735 Changing model to gpt-5 breaks the tool

#784 make markdown live vertical_overflow and refresh_interval configurable

#782 Function call dispatcher fails when extra JSON arguments are provided

#771 sgpt fails to start: `click` is imported but not declared in dependencies

#778 Feature Request: Make `markdown_live` `vertical_overflow` Configurable

#780 Add support for OpenAI SDK `extra_body` parameter in sgptrc

#486 Feature request: math formatting

#776 Multiple tool_calls Aggregation Produces Corrupted Output

#769 📝 Integration Proposal: CAJAL — Local Scientific Paper Model for ShellGPT

#749 Integration with ollama is failing

#757 gpt-5-mini not working

#767 `environment variable` wanted in `.sgptrc`

#758 it default to use a model (o4-mini-high) which hasn't existed for 8 months

#202 I asked for a ascii shrug emoji as sgpt shrugged hard.

#659 UnicodeEncodeError with Emoji Output on Windows

#748 Since update, Issu with function calling

#754 feat request: Document secure runtime injection for OPENAI_API_KEY via password manager CLI

#741 <think> tag will lead to "IndexError: list index out of range"

#689 All REPL Instances use same history

#751 Suggestion: support `.config/shell_gpt/sgptrc`

#641 shell gpt doesnt run at all