v0.9.0

What's new?

🐛 Bug fixes

• provide fallback value when option is missing in config (in commit #5ffd1b7)
• declutter the config using defaults (in commit #b0efaea)
• decrease max_tokens to 1024 (in commit #807f464)
• do not error out when pressing ctrl+c (in commit #07f0222)
• use gemini-1.5-flash by default (in commit #ebbca16)
• increase max tokens to 8k (in commit #0a4fc1e)
• quotes on commandline (in commit #427b83d)
• make gpt-4 the default model (in commit #30397c6)
• retrieve manpage for fish-native programs (in commit #cf4d4b4)
• improve responses from llama3-7b (in commit #cba7637)
• detect os when running on osx (in commit #e501c37)
• installation from github (in commit #43c0d83)
• python build (in commit #4a8e4ea)
• check if file is readable before opening it (in commit #18bff72)
• syntax error (in commit #649986f)
• strip additional characters when detecting file names (in commit #438b181)
• improve autocompletion (in commit #6466865)
• autocompletion (in commit #0585d15)
• rerun the command using python subprocess (in commit #646bd71)
• quote output from llm (in commit #af603b1)
• bind to ctrl+p (in commit #e362c2d)

🌟 New features and improvements

• redact sensitive information (in commit #7b160fc)
• add support for mistral (in commit #5655054)
• huggingface provider (in commit #6b7d8ec)
• add commandline history to completion prompt (in commit #6ee0e7d)
• display completions in fzf (in commit #e54cda6)
• include manpage in request (in commit #6e0faed)
• switch between contexts using a command (in commit #93405a7)
• supporting logging on containers and os x (in commit #343cc6b)
• add temperature for gemini (in commit #14bf669)
• support for gemini (in commit #67c4754)
• use virtual environment (in commit #ac61007)
• change temperature (in commit #0bd9b7d)
• explain commands in a different language (in commit #b2eecbc)
• add os to system prompt (in commit #1258cf0)
• status emoji (in commit #73c00d3)
• submit file contents when codifying (in commit #bfb7443)
• azure openai (in commit #069a232)
• autocomplete (in commit #e3baa57)
• optimise prompts for llama (in commit #cd503cc)
• use chat completions api (in commit #e38eed5)

⬆ Dependency updates

• bump google-generativeai from 0.5.2 to 0.7.1 (in commit #42098c7)
• bump docker/build-push-action from 5 to 6 (in commit #4877499)
• bump openai from 1.27.0 to 1.35.7 (in commit #a54420e)
• bump openai from 1.25.2 to 1.27.0 (in commit #b4c63cc)
• bump openai from 1.23.2 to 1.25.2 (in commit #f24078b)
• bump google-generativeai from 0.5.0 to 0.5.2 (in commit #a86e2bd)
• bump openai from 1.17.1 to 1.23.2 (in commit #c7c01d8)
• bump google-generativeai from 0.4.1 to 0.5.0 (in commit #cdfb1c0)
• bump openai from 1.16.2 to 1.17.1 (in commit #e591aa4)
• bump actions/stale from 5 to 9 (in commit #fca33b6)