docs: add ollama local server guide (#6)

README.md

@@ -26,6 +26,8 @@ export OPENAI_API_KEY=sk-...
 python -m first_commit_ai "Explain Git remotes like a shop dad."
 ```
 
+For a local walkthrough, see [docs/ollama.md](docs/ollama.md).
+
 ## What you just built
 
 | File | Job |

docs/ollama.md

@@ -0,0 +1,58 @@
+# Local Ollama Walkthrough
+
+Run `first-commit-ai` against a local Ollama server without paying for an API key.
+
+## 1. Install and Start Ollama
+
+1. Download Ollama from [ollama.com](https://ollama.com) and install it.
+2. Start the Ollama service:
+   ```bash
+   ollama serve
+   ```
+3. Pull a lightweight model:
+   ```bash
+   ollama pull llama3.2:1b
+   ```
+
+## 2. Configure Environment Variables
+
+Point the CLI at your local Ollama endpoint:
+
+**Linux / macOS:**
+```bash
+export OPENAI_BASE_URL="http://localhost:11434/v1"
+export OPENAI_API_KEY="ollama"
+export OPENAI_MODEL="llama3.2:1b"
+```
+
+**Windows (Command Prompt):**
+```cmd
+set OPENAI_BASE_URL=http://localhost:11434/v1
+set OPENAI_API_KEY=ollama
+set OPENAI_MODEL=llama3.2:1b
+```
+
+**Windows (PowerShell):**
+```powershell
+$env:OPENAI_BASE_URL="http://localhost:11434/v1"
+$env:OPENAI_API_KEY="ollama"
+$env:OPENAI_MODEL="llama3.2:1b"
+```
+
+## 3. Run a Live Prompt
+
+Run a real prompt without the `--mock` flag:
+```bash
+python -m first_commit_ai "Why is the sky blue?"
+```
+
+## Troubleshooting: Server Down
+
+If Ollama is not running or unreachable at the target port, the command will exit with a connection error:
+```text
+error: Connection error. Ensure Ollama is running (`ollama serve`) and reachable at http://localhost:11434.
+```
+Check that the service is running and accessible via:
+```bash
+curl http://localhost:11434/api/tags
+```