Local #1

zh30 · 2025-11-05T10:38:44Z

No description provided.

- Fix hardcoded 1024 token limit, allow selecting longer models via config - Persist selected model path in user's cache directory for reuse - Warn when no models available instead of failing - Improve error messages related to model selection

- Delay EOS token sampling until meaningful text is generated - Skip commit when diff is empty instead of failing - Handle initial commit case where no parent exists

Introduce a GitHub Actions workflow that builds release binaries for macOS (Intel & Apple Silicon), Linux (x86_64 & ARM64) and Windows. Update the release workflow to download those artifacts, compute SHA256 checksums, patch `git-ca.rb` with the new version and bottle checksums, and push the updated formula to the `homebrew-tap` repository. Refresh documentation (`CLAUDE.md`, `DEPLOY.md`, `HOMEBREW.md`, `README.md`) to describe the multi‑platform releases and Homebrew installation steps. Modify `git-ca.rb` to include a `bottle` block and install the pre‑built binary instead of building from source.

changeset-bot · 2025-11-05T10:38:47Z

⚠️ No Changeset found

Latest commit: ff14db5

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

gemini-code-assist · 2025-11-05T10:39:38Z

Summary of Changes

Hello @zh30, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces a significant architectural shift for the Git Commit Analyzer, migrating from an external Ollama dependency to a fully local llama.cpp inference engine. This change aims to improve performance, reduce external dependencies, and enhance user privacy by enabling offline AI-powered commit message generation directly within the CLI. The refactor touches core AI interaction, dependency management, and updates all relevant documentation and installation scripts to reflect this new, self-contained approach.

Highlights

AI Backend Migration: The project has transitioned from using an external Ollama API for AI inference to directly integrating with llama.cpp for local GGUF model inference. This enhances privacy and enables offline functionality.
Core Logic Refactor: A new file, src/llama.rs, has been introduced to manage llama.cpp sessions. src/main.rs underwent a significant refactor to integrate the new inference engine, including robust model selection, download, and fallback mechanisms for commit message generation.
Dependency Overhaul: The project's dependencies have been updated to reflect the new AI backend. reqwest and serde_json have been replaced with llama-cpp-sys-2, hf-hub, and rand in Cargo.toml and Cargo.lock.
Documentation & Installation Update: All user-facing and internal documentation files (README*.md, AGENTS.md, CLAUDE.md, DEPLOY.md, HOMEBREW.md, INSTALL.md) have been thoroughly revised to reflect the new architecture, installation methods (especially Homebrew with pre-built binaries), and local model management.
Homebrew Formula Enhancement: The git-ca.rb Homebrew formula has been updated to support multi-platform pre-built binaries (bottles), simplifying the installation process for macOS and Linux users by eliminating the need for local compilation.

Ignored Files

Ignored by pattern: .github/workflows/** (2)
- .github/workflows/build-binaries.yml
- .github/workflows/release.yml

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This is a significant and impressive refactoring of the project. The move from an Ollama dependency to a self-contained binary using llama.cpp is a major architectural improvement that will greatly enhance the user experience. The accompanying updates to the documentation across all languages, the release process, and the Homebrew formula are thorough and well-executed. The new code is well-structured, with robust features like model auto-downloading, deterministic fallbacks, and a doctor command for diagnostics. I have a couple of minor suggestions to improve Conventional Commits compliance and for general cleanup, but overall this is excellent work.

gemini-code-assist · 2025-11-05T10:41:32Z

Cargo.toml

+hf-hub = { version = "0.4.3", default-features = false, features = ["ureq", "native-tls"] }
+rand = "0.9"
+
+# Test comment


This comment appears to be a leftover from development or testing. It should be removed to keep the Cargo.toml file clean.

gemini-code-assist · 2025-11-05T10:41:33Z

src/main.rs

+        match lower.as_bytes().get(commit_type.len()) {
+            Some(b'(') | Some(b':') => true,
+            _ => false,
+        }


The function is_commit_subject (and parse_commit_subject which has similar logic) doesn't handle the ! indicator for breaking changes as specified by the Conventional Commits standard (e.g., feat!: ...). To fully support the spec, this should be recognized as a valid format.

Suggested change

match lower.as_bytes().get(commit_type.len()) {

Some(b'(') | Some(b':') => true,

_ => false,

}

match lower.as_bytes().get(commit_type.len()) {

Some(b'(') | Some(b':') | Some(b'!') => true,

_ => false,

}

zh30 added 14 commits October 16, 2025 19:19

Refactor code structure for improved readability and maintainability

cb7e3b1

Prevent early EOS token generation and handle empty diffs

ad5b09b

- Delay EOS token sampling until meaningful text is generated - Skip commit when diff is empty instead of failing - Handle initial commit case where no parent exists

feat(content): add claude api documentation

cc428e1

feat(main): refactor main function

7dba16d

feat(api): add llama cpp bindings

de4ee51

feat(demo): add final test documentation

587ddff

docs(docs-demo-final): update Docs and more documentation

496bc50

feat(tests): add detailed test documentation

72dea7d

docs(docs-enhanced-test): update Docs and Enhanced Test documentation

8d76aaf

feat: add final body test

c9ed670

feat: add debug body test documentation

0b9cc39

refactor(prompt): update task instruction format and error handling

ed44a94

gemini-code-assist bot reviewed Nov 5, 2025

View reviewed changes

zh30 added 2 commits November 5, 2025 18:47

fix(deploy): update windows build instructions

56bd575

feat(deploy): add macOS and Windows build instructions

ff14db5

zh30 merged commit 5b0ebac into main Nov 5, 2025
3 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

Local #1

Local #1

Uh oh!

zh30 commented Nov 5, 2025

Uh oh!

changeset-bot bot commented Nov 5, 2025 •

edited

Loading

Uh oh!

gemini-code-assist bot commented Nov 5, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

gemini-code-assist bot Nov 5, 2025

Uh oh!

gemini-code-assist bot Nov 5, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Local #1

Local #1

Uh oh!

Conversation

zh30 commented Nov 5, 2025

Uh oh!

changeset-bot bot commented Nov 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

⚠️ No Changeset found

Uh oh!

gemini-code-assist bot commented Nov 5, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist bot Nov 5, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Nov 5, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

changeset-bot bot commented Nov 5, 2025 •

edited

Loading