GCM - Git Commit Message Generator

GCM (Git Commit Message Generator) is a cross-platform tool designed to help developers craft professional and consistent Git commit messages by leveraging AI providers such as OpenAI, Codex, Claude, OpenRouter, and Ollama, combined with real-time git status and git diff analysis.

🚀 Key Features

AI-Powered Commit Message Generation
Uses OpenAI, Codex, Claude, OpenRouter, or Ollama to generate concise and detailed commit messages based on actual changes detected in the repository.
Environment Detection with Emoji
Automatically identifies the operating system or terminal environment (Windows 🪟, MacOS 🍎, Linux 🐧, CYGWIN 🪟) and includes an emoji in the commit header for visual reference.
File Change Classification
Classifies changes into Add 🆕, Change 📝, Delete 🗑️ categories with clear summaries.
Diff Summary Analysis
Performs lightweight analysis of git diff output to summarize insertions, deletions, and modifications.
Prompt Templates via Config File
Allows prompt customization through the config.yml file without modifying the source code. Multiple styles or languages can be supported.
Commit Numbering and Timestamp
Appends a unique commit number (e.g., 000,057) and precise timestamp (YYYY-MM-DD HH:MM:SS.mss) to each commit for easy tracking.
Token Usage and Execution Time
Displays real-time statistics including token usage (Prompt 📝, Response 💬, Total 🧮) and the time spent on AI response generation ⏱️.
Multi-provider Orchestration
Can combine generators, judges, and refiners across Ollama, OpenRouter, OpenAI, Codex, and Claude, depending on which credentials are available.
Structured History for Quality Analysis
In addition to the human-readable history log, GCM can persist a JSON Lines audit trail with the selected provider, displayed candidates, judge/refiner roles, and final message.
Optional Location Metadata
Location lookup is disabled by default and can be enabled with include_location: true if you want it in the commit footer.
Dual Terminal and OS Compatibility
Runs smoothly on CMD, PowerShell, Cygwin Bash, Git Bash, Linux, and MacOS terminals.

📦 Project Structure (Visual Placeholder)

GCM/
├── apis/
│   ├── OpenRouter/
│   │   ├── query_model.py
│   │   ├── secret.bash
│   │   └── secret.bat
│   ├── OpenAI/
│   │   ├── query_model.py
│   │   ├── secret.bash
│   │   └── secret.bat
│   ├── Codex/
│   │   ├── query_model.py
│   │   ├── secret.bash
│   │   └── secret.bat
│   ├── Claude/
│   │   └── query_model.py
│   └── Ollama/
│       ├── query_model.py
│       ├── secret.bash
│       └── secret.bat
├── gcm.py
├── gcm.bat
├── install.bat
├── install.bash
├── run.bat
├── run.bash
├── test.bat
├── test.bash
├── config.yml
├── requirements.txt
├── version.cfg
├── version.py
├── version.txt
└── utils.py

📸

⚙️ How It Works

Prepare the Repository: Ensure you have a valid .git directory.
Run the Script: Execute run.bat on Windows or run.bash on Unix-like systems.
Review the Suggested Commit: The AI proposes a message based on your staged or unstaged changes.
Confirm and Commit: You have the option to confirm or cancel before the actual commit is made.
Optional Staging: If nothing is staged yet, GCM asks before running git add ..

📸 With Ollama:

🔑 Configuration (config.yml example)

max_tokens: 600
use_confirmation: true
save_history: true
history_path: ~/.gcm_history.log
history_json_path: ~/.gcm_history.jsonl
include_location: false
max_characters: 500
suggested_messages: 3
provider_strategy: auto
generator_count: 3
enable_judge: true
enable_refiner: false

providers:
  ollama:
    enabled: true
    priority: 30
    roles: ["generate"]
  openrouter:
    enabled: true
    priority: 40
    roles: ["generate"]
  openai:
    enabled: true
    priority: 70
    roles: ["generate", "judge", "refine"]
  claude:
    enabled: true
    priority: 95
    roles: ["generate", "judge", "refine"]
  codex:
    enabled: true
    priority: 100
    roles: ["generate", "judge", "refine"]

emojis:
  header: "🔀"
  add: "🆕"
  change: "📝"
  delete: "🗑️"
  info: "ℹ️"
  summary: "🎯"
  windows: "🪟"
  macos: "🍎"
  linux: "🐧"

prompt_template: |
  You are an expert software engineer and Git practitioner. Based on the following Git status summary, generate a clear and complete Git commit message with the following structure:

  1. A concise summary line using the Conventional Commits format (e.g., feat:, fix:, chore:, refactor:).
  2. A short descriptive paragraph explaining what was added, changed, or removed and why.
  3. Optionally, include a brief explanation or rationale if it helps clarify the purpose of the changes.
  4. The total length of the commit message must not exceed 300 characters. If necessary, shorten the content but ensure that every sentence or idea is complete and not cut off mid-word or mid-phrase.
  5. Never leave a paragraph, sentence, or idea incomplete. Avoid unfinished sentences.
  6. Do not use markdown, use plain text.

  Changes:
  {changes}

  Diff summary:
  {diff}

history_json_path stores one JSON object per committed message. This makes it easier to compare acceptance and quality across providers such as Ollama, OpenRouter, OpenAI, Codex, and Claude.

When history_json_path points to a shared file under ~, the report aggregates commits from every repository where GCM was executed. The history report now includes a Project column before Provider so that shared history remains clear without losing cross-project visibility.

Provider availability is gated by both config.yml and environment variables:

OpenAI requires OPENAI_API_KEY
Codex requires CODEX_API_KEY
Claude requires ANTHROPIC_API_KEY
OpenRouter requires OPENROUTER_API_KEY
Ollama requires OLLAMA_MODEL

The root load_secrets.bat and load_secrets.bash loaders now source provider-specific secret files for OpenRouter, OpenAI, Codex, Claude, and Ollama.

For convenience, apis/Codex/secret.bat and apis/Codex/secret.bash default CODEX_MODEL to gpt-5-codex and reuse OPENAI_API_KEY as CODEX_API_KEY when a dedicated Codex key is not set. This makes it easy to test Codex with the current setup while still allowing an explicit CODEX_API_KEY override.

🛠 Installation

Use the provided install.bat (🪟Windows) or install.bash (🐧Linux / 🍎MacOS) to set up the environment.
Virtual environments are supported (.venv).
Creates aliases or shortcuts for seamless execution.
📸 Part 1/2
📸 Part 2/2

✅ Running Tests

Run the bundled test launcher for your shell:

test.bat on Windows CMD or PowerShell
./test.bash on Linux, macOS, Cygwin, or Git Bash

Both scripts execute the same suite:

python -m unittest discover -s tests -q

🌐 Supported Models

🤖 Ollama: Supports models like 🧠 llama3, gemma, or any local model served by Ollama.
- Local Ollama usage does not require an API key. Set OLLAMA_MODEL and ensure the Ollama host is reachable.
- Environment Variable: OLLAMA_MODEL
  - To list available models:
    - 🪟 Windows (CMD):
      - 💻 Commands:
        
        cd apis\Ollama list.bat
      - 👀 Example:
        
        C:\Users\nelbren\GCM\apis\Ollama>list.bat List models: ------------ deepseek-coder:33b qwen2.5:72b gemma3:27b deepseek-r1:32b hhao/qwen2.5-coder-tools:32b hhao/qwen2.5-coder-tools:0.5b stable-code:3b-code-q4_0 codegemma:2b gemma:2b
    - 🪟 Windows (PowerShell):
      - 💻 Commands:
        cd apis\Ollama .\run.bat
    - 🐧 Linux | 🍎 macOS | 🪟 Windows (Cygwin / Git Bash):
      - 💻 Commands:
        cd apis/Ollama ./list.bash

OpenAI: `gpt-3.5-turbo`, `gpt-4o`, or configurable models.

Environment Variable: `MODEL_TIER`

cheap
- OPENAI_MODEL=gpt-3.5-turbo
premium
- OPENAI_MODEL=gpt-4o

To list available models:

🪟 Windows (CMD):

💻 Commands:
```
cd apis\OpenAI
list.bat
```
👀 Example:

C:\Users\nelbren\GCM\apis\OpenAI>list.bat
List models:
------------
gpt-4-0613
gpt-4
gpt-3.5-turbo
o4-mini-deep-research-2025-06-26
codex-mini-latest
gpt-4o-realtime-preview-2025-06-03
gpt-4o-audio-preview-2025-06-03
o4-mini-deep-research
davinci-002
babbage-002
gpt-3.5-turbo-instruct
gpt-3.5-turbo-instruct-0914
dall-e-3
dall-e-2
gpt-4-1106-preview
gpt-3.5-turbo-1106
tts-1-hd
tts-1-1106
tts-1-hd-1106
text-embedding-3-small
text-embedding-3-large
gpt-4-0125-preview
gpt-4-turbo-preview
gpt-3.5-turbo-0125
gpt-4-turbo
gpt-4-turbo-2024-04-09
gpt-4o
gpt-4o-2024-05-13
gpt-4o-mini-2024-07-18
gpt-4o-mini
gpt-4o-2024-08-06
chatgpt-4o-latest
o1-preview-2024-09-12
o1-preview
o1-mini-2024-09-12
o1-mini
gpt-4o-realtime-preview-2024-10-01
gpt-4o-audio-preview-2024-10-01
gpt-4o-audio-preview
gpt-4o-realtime-preview
omni-moderation-latest
omni-moderation-2024-09-26
gpt-4o-realtime-preview-2024-12-17
gpt-4o-audio-preview-2024-12-17
gpt-4o-mini-realtime-preview-2024-12-17
gpt-4o-mini-audio-preview-2024-12-17
o1-2024-12-17
o1
gpt-4o-mini-realtime-preview
gpt-4o-mini-audio-preview
o3-mini
o3-mini-2025-01-31
gpt-4o-2024-11-20
gpt-4.5-preview
gpt-4.5-preview-2025-02-27
gpt-4o-search-preview-2025-03-11
gpt-4o-search-preview
gpt-4o-mini-search-preview-2025-03-11
gpt-4o-mini-search-preview
gpt-4o-transcribe
gpt-4o-mini-transcribe
o1-pro-2025-03-19
o1-pro
gpt-4o-mini-tts
o4-mini-2025-04-16
o4-mini
gpt-4.1-2025-04-14
gpt-4.1
gpt-4.1-mini-2025-04-14
gpt-4.1-mini
gpt-4.1-nano-2025-04-14
gpt-4.1-nano
gpt-image-1
gpt-3.5-turbo-16k
tts-1
whisper-1
text-embedding-ada-002

🪟 Windows (PowerShell):
- 💻 Commands:
```
cd apis\OpenAI
.\run.bat
```
🐧 Linux | 🍎 macOS | 🪟 Windows (Cygwin / Git Bash):
- 💻 Commands:
```
cd apis/OpenAI
./list.bash
```

Environment Variable: OPENAI_MODEL
- Select specific model from OpenAI

Codex: gpt-5-codex or any compatible Codex-capable model.
- Environment Variables: CODEX_API_KEY, CODEX_MODEL
- By default, CODEX_MODEL falls back to gpt-5-codex.
- If CODEX_API_KEY is unset but OPENAI_API_KEY is available, the bundled Codex secret scripts reuse the OpenAI key so Codex can be tested without duplicating credentials.
Claude: configurable Anthropic models.
- Environment Variables: ANTHROPIC_API_KEY, CLAUDE_MODEL
- CLAUDE_MODEL defaults to claude-3-5-sonnet-latest.

OpenRouter: any free models.

Environment Variable: `OPENROUTER_MODEL`

FreeAll

Selects any available free model, without prioritization. Broadest selection.
FreeTop

Selects the top-rated free models based on community feedback and performance.
FreeCtxMax

Prioritizes free models with the largest context window available for longer prompts.
FreeSmart

Dynamically selects the most balanced free model considering speed, context, and quality.

To list available models:

🪟 Windows (CMD):

💻 Commands:
```
cd apis\OpenRoute
list.bat
```
👀 Example:

C:\Users\nelbren\GCM\apis\OpenRouter>list.bat
Free models:
------------
tencent/hunyuan-a13b-instruct:free
tngtech/deepseek-r1t2-chimera:free
openrouter/cypher-alpha:free
mistralai/mistral-small-3.2-24b-instruct:free
moonshotai/kimi-dev-72b:free
deepseek/deepseek-r1-0528-qwen3-8b:free
deepseek/deepseek-r1-0528:free
sarvamai/sarvam-m:free
mistralai/devstral-small:free
google/gemma-3n-e4b-it:free
qwen/qwen3-30b-a3b:free
qwen/qwen3-8b:free
qwen/qwen3-14b:free
qwen/qwen3-32b:free
qwen/qwen3-235b-a22b:free
tngtech/deepseek-r1t-chimera:free
microsoft/mai-ds-r1:free
thudm/glm-z1-32b:free
thudm/glm-4-32b:free
shisa-ai/shisa-v2-llama3.3-70b:free
arliai/qwq-32b-arliai-rpr-v1:free
agentica-org/deepcoder-14b-preview:free
moonshotai/kimi-vl-a3b-thinking:free
nvidia/llama-3.3-nemotron-super-49b-v1:free
meta-llama/llama-4-maverick:free
meta-llama/llama-4-scout:free
deepseek/deepseek-v3-base:free
google/gemini-2.5-pro-exp-03-25
qwen/qwen2.5-vl-32b-instruct:free
deepseek/deepseek-chat-v3-0324:free
featherless/qwerky-72b:free
mistralai/mistral-small-3.1-24b-instruct:free
google/gemma-3-4b-it:free
google/gemma-3-12b-it:free
rekaai/reka-flash-3:free
google/gemma-3-27b-it:free
qwen/qwq-32b:free
nousresearch/deephermes-3-llama-3-8b-preview:free
cognitivecomputations/dolphin3.0-r1-mistral-24b:free
cognitivecomputations/dolphin3.0-mistral-24b:free
qwen/qwen2.5-vl-72b-instruct:free
mistralai/mistral-small-24b-instruct-2501:free
deepseek/deepseek-r1-distill-qwen-14b:free
deepseek/deepseek-r1-distill-llama-70b:free
deepseek/deepseek-r1:free
deepseek/deepseek-chat:free
google/gemini-2.0-flash-exp:free
meta-llama/llama-3.3-70b-instruct:free
qwen/qwen-2.5-coder-32b-instruct:free
meta-llama/llama-3.2-11b-vision-instruct:free
qwen/qwen-2.5-72b-instruct:free
mistralai/mistral-nemo:free
google/gemma-2-9b-it:free
mistralai/mistral-7b-instruct:free

🪟 Windows (PowerShell):
- 💻 Commands:
```
cd apis\OpenRoute
.\run.bat
```
🐧 Linux | 🍎 macOS | 🪟 Windows (Cygwin / Git Bash):
- 💻 Commands:
```
cd apis/OpenRoute
./list.bash
```

🤖 Example Commit Message Output

📸 With OpenAI:

🔁 Round-robin and OpenRouter support

📸 With suggested_messages: 3

📊 History Analytics

If save_history: true and history_json_path is enabled, you can inspect provider acceptance and average response times with:

python report_history.py

The report is designed for terminal usage and uses Rich tables with color when the dependency is installed. Recommended reading order:

Accept/Shown: relative acceptance of a provider against the times it was actually shown as a candidate.
Accept/Commits: overall share of final selected messages across all analyzed commits.
Avg Selected: average elapsed time when that provider ended up selected.
Avg Shown: average elapsed time across all displayed messages from that provider.
Gen, Judge, Refine: how often the provider was used in each orchestration role.

This gives you a concrete way to compare whether Codex, Claude, OpenAI, Ollama, or OpenRouter are producing the best tradeoff between quality and latency.

👨‍💻 Contributing

Feedback, suggestions, and improvements are welcome. Feel free to open an issue or submit a pull request.

🙏 Credits

Developed by 🧑‍💻 Nelbren with 🤖 AI assistance from Aren 😎 (ChatGPT).
Prompt engineering, code generation, emoji madness, and technical fine-tuning powered by Aren.

📄 License

MIT License (or specify your preferred license here)

📆 2026-04-18 23:41:50 🪟

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

GCM - Git Commit Message Generator

🚀 Key Features

📦 Project Structure (Visual Placeholder)

⚙️ How It Works

🔑 Configuration (config.yml example)

🛠 Installation

✅ Running Tests

🌐 Supported Models

🤖 Ollama: Supports models like 🧠 `llama3`, `gemma`, or any local model served by Ollama.

Environment Variable: `OLLAMA_MODEL`

To list available models:

OpenAI: `gpt-3.5-turbo`, `gpt-4o`, or configurable models.

Environment Variable: `MODEL_TIER`

`cheap`

OPENAI_MODEL=gpt-3.5-turbo

`premium`

OPENAI_MODEL=gpt-4o

To list available models:

Environment Variable: `OPENAI_MODEL`

Select specific model from OpenAI

Codex: `gpt-5-codex` or any compatible Codex-capable model.

Environment Variables: `CODEX_API_KEY`, `CODEX_MODEL`

Claude: configurable Anthropic models.

Environment Variables: `ANTHROPIC_API_KEY`, `CLAUDE_MODEL`

OpenRouter: any free models.

Environment Variable: `OPENROUTER_MODEL`

`FreeAll`

`FreeTop`

`FreeCtxMax`

`FreeSmart`

To list available models:

🤖 Example Commit Message Output

🔁 Round-robin and OpenRouter support

📊 History Analytics

👨‍💻 Contributing

🙏 Credits

📄 License

FilesExpand file tree

README.md

Latest commit

History

README.md

File metadata and controls

GCM - Git Commit Message Generator

🚀 Key Features

📦 Project Structure (Visual Placeholder)

⚙️ How It Works

🔑 Configuration (config.yml example)

🛠 Installation

✅ Running Tests

🌐 Supported Models

🤖 Ollama: Supports models like 🧠 llama3, gemma, or any local model served by Ollama.

Environment Variable: OLLAMA_MODEL

To list available models:

OpenAI: gpt-3.5-turbo, gpt-4o, or configurable models.

Environment Variable: MODEL_TIER

cheap

OPENAI_MODEL=gpt-3.5-turbo

premium

OPENAI_MODEL=gpt-4o

To list available models:

Environment Variable: OPENAI_MODEL

Select specific model from OpenAI

Codex: gpt-5-codex or any compatible Codex-capable model.

Environment Variables: CODEX_API_KEY, CODEX_MODEL

Claude: configurable Anthropic models.

Environment Variables: ANTHROPIC_API_KEY, CLAUDE_MODEL

OpenRouter: any free models.

Environment Variable: OPENROUTER_MODEL

FreeAll

FreeTop

FreeCtxMax

FreeSmart

To list available models:

🤖 Example Commit Message Output

🔁 Round-robin and OpenRouter support

📊 History Analytics

👨‍💻 Contributing

🙏 Credits

📄 License

🤖 Ollama: Supports models like 🧠 `llama3`, `gemma`, or any local model served by Ollama.

Environment Variable: `OLLAMA_MODEL`

OpenAI: `gpt-3.5-turbo`, `gpt-4o`, or configurable models.

Environment Variable: `MODEL_TIER`

`cheap`

`premium`

Environment Variable: `OPENAI_MODEL`

Codex: `gpt-5-codex` or any compatible Codex-capable model.

Environment Variables: `CODEX_API_KEY`, `CODEX_MODEL`

Environment Variables: `ANTHROPIC_API_KEY`, `CLAUDE_MODEL`

Environment Variable: `OPENROUTER_MODEL`

`FreeAll`

`FreeTop`

`FreeCtxMax`

`FreeSmart`