1 # Changelog
2
3 ## 1.0.4
4
5 This release tightens up the terminal experience and finishes a few release-facing cleanup items.
6
7 ### What changed
8
9 - Added bold rich-text rendering in the TUI for assistant replies, so `**text**` now displays with terminal styling instead of raw markdown markers
10 - Refreshed the bundled skill metadata to follow the current Agent Skills `SKILL.md` format
11 - Synced the crate release metadata for the `1.0.4` cut
12
13 ## 1.0.3
14
15 This is the cleanup release for the editor-side startup problems.
16
17 ### What changed
18
19 - Fixed ACP sessions failing on the first real prompt because the server claimed the model was ready before anything had actually been loaded
20 - Changed ACP startup so the default model loads lazily on the first non-slash prompt instead of pretending it is already in memory
21 - Kept `initialize` and `session/new` lightweight while still sending proper progress updates once model loading begins
22 - Updated the Onde integration to `1.0.0`
23 - Removed a few dependencies we were no longer using
24
25 ## 1.0.2
26
27 This release was supposed to fix the ACP auth breakage. It did fix the stdout pollution problem, but it turned out not to be the whole story.
28
29 ### What changed
30
31 - Delayed model loading in ACP mode so startup diagnostics would not leak into protocol stdout during the auth handshake
32 - Tightened up the ACP startup path for editor integrations
33 - Refreshed some README wording while cutting `1.0.2`
34
35 ## 1.0.1
36
37 The first patch after `1.0.0` was mostly about making model loading and model switching feel less opaque.
38
39 ### What changed
40
41 - Added ToolCall-based progress UI for startup model loading and downloading
42 - Improved model-switch progress reporting in ACP clients
43 - Fixed model-load error handling so failed switches did not leave the UI in a weird state
44 - Cleaned up a few status messages and docs while the release was going out
45
46 ## 1.0.0 (2026)
47
48 siGit Code has been living in real smbCloud repos for a while now. At some point it stopped feeling like an experiment, so we called it 1.0.
49
50 ### What this release is
51
52 siGit Code is a local coding agent. It runs a quantized model on your machine, talks to editors over ACP, and can read files, run commands, fetch web pages, and write code without sending your project to a hosted API.
53
54 You can install it with Cargo, pip, npm, or Homebrew and use it like any other tool on your machine.
55
56 ### What shipped in 1.0
57
58 #### Editor integration
59
60 This is the core of the project.
61
62 Zed and VS Code can talk to siGit Code over ACP. Multi-turn sessions work. Tool calling works. Session forking works. Working-directory context works. That was the original goal, and it feels solid now.
63
64 #### Terminal UI
65
66 The terminal UI started as a side quest and turned out to be useful. You get a full-screen ratatui chat, streaming tokens, a spinner while the model is busy, and a model picker you can open in the middle of a session.
67
68 It runs on macOS and Linux. Windows gets ACP and editor mode for now. The Windows terminal UI is still unfinished.
69
70 #### Tool calling
71
72 This is the part that makes siGit Code feel like an agent instead of a chat box. The loop can run up to 10 rounds per message.
73
74 Available tools:
75
76 - `read_file` / `write_file` / `delete_file`
77 - `list_directory` / `search_files`
78 - `run_command`, with an optional working directory
79 - `read_website`, which fetches a URL and strips it down to readable text
80
81 The model can call a tool, inspect the result, and keep going until it has a real answer.
82
83 #### Model support
84
85 The model list ended up wider than we expected for 1.0:
86
87 - Qwen 3 1.7B, 4B, 8B, and 14B
88 - Qwen 2.5 1.5B and 3B
89 - Qwen 2.5 Coder 1.5B, 3B, and 7B
90 - DeepSeek Coder 6.7B
91
92 They are all GGUF models and they all come from Hugging Face on first run.
93
94 Qwen 3 is the interesting one. It uses extended thinking mode. The model reasons inside `<think>...</think>` blocks before answering. The TUI strips those blocks out and renders them dimmed above the reply, so you can see what happened without turning the whole conversation into noise.
95
96 The 8B model is the desktop default. Mobile stays on 1.7B because iOS gives apps roughly 2 to 3 GB of memory, and we learned the hard way that 3B can blow up on an iPhone 16e.
97
98 #### Model picker
99
100 The model picker shows:
101
102 - what is already cached locally
103 - what can be downloaded
104 - which models support tool calling
105 - whether the local cache looks healthy
106
107 You can open it with `/models` in the TUI or through the editor config option. Switching models happens in the background and the UI stays alive while the download or load is in progress.
108
109 #### smbCloud context
110
111 siGit Code knows smbCloud repos better than a generic coding assistant does. It understands the difference between platform-user flows and tenant-app auth flows, how `Project`, `FrontendApp`, `AuthApp`, and GresIQ fit together, and why Next.js SSR deploys are not the same thing as the generic git-push path.
112
113 Outside smbCloud, it backs off and behaves like a normal coding agent.
114
115 #### Distribution
116
117 Distribution took an unreasonable amount of time, honestly.
118
119 There are prebuilt binaries for macOS, Linux, and Windows on both arm64 and x64 where relevant, plus install paths through Cargo, PyPI, npm, and Homebrew:
120
121 - `cargo install sigit`
122 - `pip install sigit-code`
123 - `npm install -g @smbcloud/sigit`
124 - `brew install sigit`
125
126 Getting that whole pipeline to behave across CI, crates.io, PyPI, npm, and Homebrew was basically its own project.
127
128 ### What still does not work
129
130 The Windows terminal UI is still missing.
131
132 ACP and editor mode work on Windows. The part that is still missing is the interactive full-screen terminal UI. The blocker is Unix-specific terminal handling that has not been abstracted cleanly yet.
133
134 ### Changes since 0.1.2
135
136 - Added Qwen 3 14B support
137 - Added Qwen 3 `<think>` block parsing and separate rendering in the TUI
138 - Moved all TUI code into `#[cfg(unix)]`, which fixed a pile of dead-code errors on Windows CI
139 - Added live download progress during model switches, including cancellation with Ctrl+C
140 - Added model download and loading progress in the Zed agent config panel
141 - Added an animated spinner during model switching
142 - Added Qwen 2.5 Coder 7B
143 - Added downloadable models to the picker, not just locally cached ones
144 - Made model selection persist across restarts
145 - Added session working-directory support
146 - Moved model picker logic into a platform-independent module so Windows can compile without the TUI
147 - Added the `/models N` shortcut for picking a model by number
148 - Added the `read_website` tool
149 - Improved `read_file` handling and empty-reply detection
150 - Added async tool execution
151 - Fixed CI cross-compilation for macOS, iOS, Linux, and Windows
152 - Added npm, PyPI, and Homebrew distribution
153
154 ---
155
156 *© 2026 [smbCloud](https://smbcloud.xyz/) (Splitfire AB).*