chore: prepare v0.6.0 release

- Clean up .gitignore (remove Swift/Xcode legacy entries, add Tauri entries)
- Add container autostart documentation for all platforms
- Add podman machine autostart instructions (macOS/Windows)
- Add systemd socket + lingering setup for Linux
- Configure restart: always policy in compose.yaml
- Update CLAUDE.md with TODO stub requirements
- Bump version to 0.6.0

Co-Authored-By: Claude Opus 4.5 <[email protected]>

Author: 2 people

.gitignore

@@ -1,42 +1,24 @@
-# Swift Package Manager
-.build/
-.build-*/
-.swiftpm/
-Package.resolved
-
-#Sparkle Directory
-sparkle/
-
 # Claude Code local settings
 .claude/
 
-# Xcode
-*.xcodeproj/
-*.xcworkspace/
-xcuserdata/
-DerivedData/
-*.moved-aside
-*.pbxuser
-!default.pbxuser
-*.mode1v3
-!default.mode1v3
-*.mode2v3
-!default.mode2v3
-*.perspectivev3
-!default.perspectivev3
-*.hmap
-*.ipa
-*.dSYM.zip
-*.dSYM
+# Rust/Cargo
+target/
+Cargo.lock
+!tauri-app/src-tauri/Cargo.lock
+
+# Node.js
+node_modules/
+.pnpm-store/
+
+# Tauri
+tauri-app/.svelte-kit/
+tauri-app/build/
 
 # Build artifacts
-build/
 *.o
 *.swp
 *~
-*.app/
 *.zip
-*.iconset/
 
 # macOS
 .DS_Store
@@ -50,11 +32,13 @@ build/
 *.sublime-workspace
 .vscode/
 
-# Archives
-*.xcarchive
-
 # Crash logs
 *.crash
 
 # Worktrees
 .worktrees/
+
+# Environment files
+.env
+.env.*
+!.env.example

CLAUDE.md

@@ -10,6 +10,59 @@ Claude Code Monitor is a cross-platform desktop application for monitoring Claud
 
 **Supported Platforms:** macOS, Linux, Windows
 
+
+## Planning, TODOs, and Source of Truth (Required)
+
+Claude may generate design and implementation plans (including Obra / Superpowers outputs) to reason about work, sequencing, and risk. These plans are **temporary reasoning artifacts**, not the authoritative record of unfinished work.
+
+The **codebase itself is the source of truth**.
+
+### Mandatory Inline TODO Stubs
+
+Whenever code, configuration, tests, or documentation contain missing, incomplete, deferred, or intentionally out-of-scope work, Claude **must insert TODO stubs directly into the relevant files** at the exact location where the work belongs.
+
+Rules:
+- TODOs must be **explicit, scoped, and actionable** — never vague.
+- Each TODO must clearly state:
+  - WHAT is missing or unimplemented
+  - WHERE / WHEN it applies (function, module, path, lifecycle stage)
+  - EXPECTED behavior or acceptance criteria (brief but testable)
+  - Any known constraints or pending decisions (if applicable)
+- TODOs must be written **inline in the affected file**, not only in chat or plans.
+- Do not leave conceptual or implied gaps without a corresponding TODO.
+
+Examples:
+- ❌ `// TODO: fix later`
+- ❌ `// TODO: handle edge cases`
+- ✅ `// TODO(auth): Validate JWT issuer and audience before accepting token — reject with 401 and log reason`
+- ✅ `# TODO(helm): Document required S3 values — include example values.yaml snippet`
+
+### Plans → TODOs Collapse Rule
+
+Any item identified during planning that is not fully implemented **must be converted into one or more inline TODO stubs** in the appropriate files.
+
+A planned item is considered handled only if:
+- it is fully implemented, or
+- it exists as a TODO in the correct location(s)
+
+Do not rely on external plan documents to represent unfinished work.
+
+### Default Behavior
+
+Prefer:
+- small, focused changes
+- incremental implementation
+- inline TODO stubs
+
+Over:
+- large speculative plans
+- separate backlog documents
+- deferred work tracked only in chat or markdown plans
+
+If something is not implemented, there must be a TODO exactly where it belongs.
+If it is implemented, the TODO should be removed.
+
+
 ## Repository Structure
 
 ```

compose.yaml

@@ -2,6 +2,8 @@ services:
   # OpenTelemetry Collector
   otel-collector:
     image: otel/opentelemetry-collector-contrib:latest
+    container_name: otel-collector
+    restart: always
     command: ["--config=/etc/otel-collector-config.yaml"]
     volumes:
       - ./otel-collector-config.yaml:/etc/otel-collector-config.yaml
@@ -16,6 +18,7 @@ services:
   prometheus:
     image: prom/prometheus:latest
     container_name: prometheus
+    restart: always
     ports:
       - "9090:9090"
     volumes:

docs/installation/linux.md

@@ -42,7 +42,24 @@ podman --version
 podman-compose version
 ```
 
-### 3. Install Podman Desktop (Optional)
+### 3. Enable Podman Socket for Autostart (Recommended)
+
+To ensure containers can restart automatically after reboot, enable the Podman socket:
+
+```bash
+systemctl --user enable podman.socket
+systemctl --user start podman.socket
+```
+
+**Enable lingering** so user services run even when not logged in:
+
+```bash
+sudo loginctl enable-linger $USER
+```
+
+This allows the monitoring stack to start automatically on boot.
+
+### 4. Install Podman Desktop (Optional)
 
 Podman Desktop provides a GUI for managing containers.
 

docs/installation/macos.md

@@ -20,7 +20,7 @@ Visit [podman-desktop.io](https://podman-desktop.io/) and download the macOS ins
 
 ### 2. Initialize Podman Machine
 
-After installing Podman Desktop, initialize the Podman machine:
+After installing Podman Desktop, initialize and start the Podman machine:
 
 ```bash
 podman machine init
@@ -36,6 +36,22 @@ podman compose version
 
 Both commands should return version information.
 
+### 3. Configure Autostart (Recommended)
+
+To have the Podman machine start automatically when you log in:
+
+```bash
+podman machine set --autostart
+```
+
+**Verify autostart is enabled:**
+
+```bash
+podman machine inspect | grep -i autostart
+```
+
+This ensures the monitoring stack is available immediately after login without manual intervention.
+
 ## Install the App
 
 ### Option A: Download Release (Recommended)

docs/installation/windows.md

@@ -30,6 +30,22 @@ podman machine init
 podman machine start
 ```
 
+### 3. Configure Autostart (Recommended)
+
+To have the Podman machine start automatically when Windows starts:
+
+```powershell
+podman machine set --autostart
+```
+
+**Verify autostart is enabled:**
+
+```powershell
+podman machine inspect | Select-String "autostart"
+```
+
+This ensures the monitoring stack is available immediately after boot without manual intervention.
+
 ## Install the App
 
 ### Option A: Download Release (Recommended)

docs/monitoring-stack.md

@@ -97,6 +97,29 @@ podman compose logs -f prometheus
 podman compose logs -f otel-collector
 ```
 
+## Configure Container Autostart
+
+To ensure the monitoring stack starts automatically when the Podman machine starts (or after a reboot on Linux), the containers are configured with `restart: always` in `compose.yaml`.
+
+**Verify restart policy:**
+
+```bash
+podman inspect prometheus --format '{{.HostConfig.RestartPolicy.Name}}'
+podman inspect otel-collector --format '{{.HostConfig.RestartPolicy.Name}}'
+```
+
+Both should output `always`.
+
+**Complete autostart requirements:**
+
+| Platform | Podman Machine Autostart | Container Restart Policy |
+|----------|-------------------------|-------------------------|
+| macOS | `podman machine set --autostart` | `restart: always` in compose.yaml |
+| Windows | `podman machine set --autostart` | `restart: always` in compose.yaml |
+| Linux | `systemctl --user enable podman.socket` + lingering | `restart: always` in compose.yaml |
+
+See platform-specific installation guides for detailed setup.
+
 ## Manage the Stack
 
 ### Stop the stack

tauri-app/package.json

@@ -1,6 +1,6 @@
 {
   "name": "tauri-app",
-  "version": "0.5.14",
+  "version": "0.6.0",
   "description": "",
   "type": "module",
   "scripts": {

tauri-app/src-tauri/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "claude-code-monitor"
-version = "0.5.14"
+version = "0.6.0"
 description = "Monitor Claude Code usage via Prometheus telemetry"
 authors = ["cragr"]
 edition = "2021"

tauri-app/src-tauri/tauri.conf.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://schema.tauri.app/config/2",
   "productName": "Claude Code Monitor",
-  "version": "0.5.14",
+  "version": "0.6.0",
   "identifier": "com.cragr.claudecodemonitor",
   "build": {
     "beforeDevCommand": "pnpm dev",