feat: add 4 starter configs + jarvis init --preset command

New example configs:
- deep-research.toml — multi-hop research with citations
- code-assistant.toml — orchestrator with code execution + file I/O
- scheduled-monitor.toml — persistent operative on cron schedule
- chat-simple.toml — lightweight chat, no tools

CLI:
- `jarvis init --preset <name>` installs any starter config in one command
- Presets: morning-digest-mac, morning-digest-linux, morning-digest-minimal,
  deep-research, code-assistant, scheduled-monitor, chat-simple

All configs tested live on M2 Max with Ollama + Qwen3.5 9B.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

README.md

@@ -79,19 +79,32 @@ uv run jarvis ask "What is the capital of France?"
 
 ## Starter Configs
 
-Copy a config to `~/.openjarvis/config.toml` to get started with a pre-built use case. Each config includes the model, agent, tools, and connectors you need.
-
-| Use Case | Config | What it does |
-|----------|--------|-------------|
-| **Morning Digest** | [`morning-digest-mac.toml`](configs/openjarvis/examples/morning-digest-mac.toml) | Daily spoken briefing from your email, calendar, health tracker, and news — delivered by a Jarvis-style AI voice |
-| **Morning Digest (minimal)** | [`morning-digest-minimal.toml`](configs/openjarvis/examples/morning-digest-minimal.toml) | Just Gmail + Calendar, runs on any machine |
-| **Morning Digest (Linux)** | [`morning-digest-linux.toml`](configs/openjarvis/examples/morning-digest-linux.toml) | For Linux servers with GPU |
+Install any preset with one command:
 
 ```bash
-# Example: set up Morning Digest on Mac
-cp configs/openjarvis/examples/morning-digest-mac.toml ~/.openjarvis/config.toml
+jarvis init --preset morning-digest-mac   # or any preset below
+```
+
+| Preset | Use Case | What it does |
+|--------|----------|-------------|
+| `morning-digest-mac` | Daily Briefing (Mac) | Spoken briefing from email, calendar, health, news with Jarvis voice |
+| `morning-digest-linux` | Daily Briefing (Linux) | Same, with vLLM support for GPU servers |
+| `morning-digest-minimal` | Daily Briefing (minimal) | Just Gmail + Calendar, runs on any machine |
+| `deep-research` | Research Assistant | Multi-hop research across indexed docs with citations |
+| `code-assistant` | Code Companion | Agent with code execution, file I/O, and shell access |
+| `scheduled-monitor` | Persistent Monitor | Stateful agent that runs on a schedule with memory |
+| `chat-simple` | Simple Chat | Lightweight conversation, no tools needed |
+
+```bash
+# Example: Morning Digest on Mac
+jarvis init --preset morning-digest-mac
 jarvis connect gdrive          # one OAuth flow covers Gmail, Calendar, Tasks
 jarvis digest --fresh           # generate and play your first briefing
+
+# Example: Deep Research
+jarvis init --preset deep-research
+jarvis memory index ./docs/    # index your documents
+jarvis ask "Summarize all emails about Project X"
 ```
 
 ### Built-in Agents

configs/openjarvis/examples/chat-simple.toml

@@ -0,0 +1,24 @@
+# Simple Chat — lightweight conversational AI, no tools
+# Copy to ~/.openjarvis/config.toml
+#
+# The fastest setup: just Ollama + a model.
+#
+# Usage:
+#   jarvis ask "What is quantum computing?"
+#   jarvis chat                  # interactive chat session
+#   jarvis serve                 # start API server for browser/desktop app
+
+[engine]
+default = "ollama"
+
+[intelligence]
+default_model = "qwen3.5:4b"       # Fast and lightweight
+# default_model = "qwen3.5:9b"     # Better quality
+# default_model = "llama3.1:8b"    # Alternative model
+
+[agent]
+default_agent = "simple"            # Single-turn, no tools
+
+[server]
+host = "0.0.0.0"
+port = 8000

configs/openjarvis/examples/code-assistant.toml

@@ -0,0 +1,21 @@
+# Code Assistant — agent with code execution, file I/O, and shell access
+# Copy to ~/.openjarvis/config.toml
+#
+# Usage:
+#   jarvis ask "Write a Python script that parses CSV files"
+#   jarvis ask "Read main.py and explain the architecture"
+#   jarvis ask --agent orchestrator "Find and fix the bug in test_utils.py"
+
+[engine]
+default = "ollama"
+
+[intelligence]
+default_model = "qwen3.5:9b"
+# default_model = "qwen3.5:35b"    # Better for complex code tasks
+
+[agent]
+default_agent = "orchestrator"      # Multi-turn with tool selection
+max_turns = 10
+
+[tools]
+enabled = ["code_interpreter", "file_read", "file_write", "shell_exec", "web_search", "think", "calculator"]

configs/openjarvis/examples/deep-research.toml

@@ -0,0 +1,27 @@
+# Deep Research Agent — multi-hop research across your indexed documents
+# Copy to ~/.openjarvis/config.toml
+#
+# First index your documents:
+#   jarvis memory index ./docs/
+#   jarvis memory index ~/Documents/papers/
+#
+# Then ask complex questions:
+#   jarvis ask --agent deep_research "Summarize all emails about Project X"
+#   jarvis ask --agent deep_research "What meetings did I have with Alice last month?"
+
+[engine]
+default = "ollama"
+
+[intelligence]
+default_model = "qwen3.5:9b"
+temperature = 0.3               # Low temperature for factual research
+
+[agent]
+default_agent = "deep_research"
+max_turns = 8                   # Multi-hop reasoning steps
+
+[tools]
+enabled = ["knowledge_search", "knowledge_sql", "scan_chunks", "think", "web_search"]
+
+[tools.storage]
+default_backend = "sqlite"

configs/openjarvis/examples/scheduled-monitor.toml

@@ -0,0 +1,35 @@
+# Scheduled Monitor — persistent agent that runs on a schedule
+# Copy to ~/.openjarvis/config.toml
+#
+# The operative agent maintains state across runs, making it ideal for:
+# - Daily email/inbox monitoring
+# - Recurring status checks
+# - Long-running research projects
+#
+# Setup:
+#   1. Index your data: jarvis memory index ~/Documents/
+#   2. Start the scheduler: jarvis scheduler start
+#   3. Create a task:
+#      jarvis scheduler create \
+#        --prompt "Check for new emails about Project X and update your notes" \
+#        --schedule "0 9 * * 1-5" \
+#        --agent operative \
+#        --tools "knowledge_search,knowledge_sql,memory_store,think"
+
+[engine]
+default = "ollama"
+
+[intelligence]
+default_model = "qwen3.5:9b"
+temperature = 0.3
+
+[agent]
+default_agent = "operative"
+max_turns = 20
+context_from_memory = true          # Inject relevant memory into context
+
+[tools]
+enabled = ["knowledge_search", "knowledge_sql", "scan_chunks", "memory_store", "memory_search", "think", "web_search"]
+
+[tools.storage]
+default_backend = "sqlite"

src/openjarvis/cli/init_cmd.py

@@ -264,6 +264,23 @@ def _do_download(engine: str, model: str, spec, console: Console) -> None:
     default=False,
     help="Include Morning Digest config section.",
 )
+@click.option(
+    "--preset",
+    type=click.Choice(
+        [
+            "morning-digest-mac",
+            "morning-digest-linux",
+            "morning-digest-minimal",
+            "deep-research",
+            "code-assistant",
+            "scheduled-monitor",
+            "chat-simple",
+        ],
+        case_sensitive=False,
+    ),
+    default=None,
+    help="Use a pre-built starter config instead of generating one.",
+)
 def init(
     force: bool,
     config: Optional[Path],
@@ -273,6 +290,7 @@ def init(
     skip_scan: bool = False,
     host: Optional[str] = None,
     enable_digest: bool = False,
+    preset: Optional[str] = None,
 ) -> None:
     """Detect hardware and generate ~/.openjarvis/config.toml."""
     console = Console()
@@ -284,6 +302,42 @@ def init(
         console.print("Use [bold]--force[/bold] to overwrite.")
         raise SystemExit(1)
 
+    # Handle --preset: copy a starter config and return early
+    if preset:
+
+        examples_dir = (
+            Path(__file__).resolve().parents[2]
+            / "configs"
+            / "openjarvis"
+            / "examples"
+        )
+        # Also check installed package location
+        if not examples_dir.exists():
+            examples_dir = (
+                Path(__file__).resolve().parents[3]
+                / "configs"
+                / "openjarvis"
+                / "examples"
+            )
+        preset_path = examples_dir / f"{preset}.toml"
+        if not preset_path.exists():
+            console.print(f"[red]Preset '{preset}' not found.[/red]")
+            console.print(
+                f"  Looked in: {examples_dir}"
+            )
+            raise SystemExit(1)
+        DEFAULT_CONFIG_DIR.mkdir(parents=True, exist_ok=True)
+        DEFAULT_CONFIG_PATH.write_text(preset_path.read_text())
+        console.print(
+            f"[green]Preset '{preset}' installed to "
+            f"{DEFAULT_CONFIG_PATH}[/green]"
+        )
+        console.print(
+            "\n  Edit the config to customize, then run "
+            "[bold]jarvis doctor[/bold] to verify."
+        )
+        return
+
     console.print("[bold]Detecting hardware...[/bold]")
     hw = detect_hardware()