Repository navigation
Local-first AI: any local model server, cloud provider opt-in - #11
Merged
Merged
Conversation
- New "local" provider for OpenAI-style servers (llama.cpp, LM Studio, Jan, LocalAI, vLLM). Finds a running server on the usual ports and uses its first model. - Anthropic is off until "enabled": true. An API key in the environment alone no longer turns it on; keys typed into older configs still do. - Model servers outside this machine and the local network are refused unless "allow_remote_ai" is set, before any request is sent. - All providers share one prompt with the safety rules (Ollama's lacked them), and get the distro family so derivatives get the right package manager. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Clishe's AI now stays on your machine unless you deliberately choose otherwise.
What changes
localprovider talks to anything with an OpenAI-style API: llama.cpp'sllama-server, LM Studio, Jan, LocalAI, vLLM. With no config, it looks on the usual ports (8080, 1234, 1337, 8000) and uses the first model the server lists."anthropic": {"enabled": true}. AnANTHROPIC_API_KEYset for some other tool no longer makes Clishe use the cloud. Older configs with a key typed into the config file are treated as opted in, so nobody's setup breaks.ollamaorlocalpoints at a host outside this computer or the local network, Clishe refuses before sending anything, and explains how to allow it ("allow_remote_ai": true). LAN addresses (a home server) and.localnames are allowed.ID_LIKE) and its package manager, so Mint and Pop!_OS getapt.localautomatically. It's inserted afterollama.Checked
--action resolvefound it, used its model, and returned the command, while a fakeANTHROPIC_API_KEYin the environment was ignored.README, SECURITY.md, CONTRIBUTING.md and the CHANGELOG are updated.
🤖 Generated with Claude Code