Xcode IDE and Xcode MCP support + Insane refactor of tool/workflow management, discovery and run-time visibility

.claude/agents/xcodebuild-mcp-qa-tester.md

@@ -76,7 +76,7 @@ After testing `list_sims` tool, update the report:
 ## Detailed Test Results
 
 ### Tool: list_sims ✅ PASSED
-**Command:** `npx reloaderoo@latest inspect call-tool list_sims --params '{}' -- node build/index.js`
+**Command:** `npx reloaderoo@latest inspect call-tool list_sims --params '{}' -- node build/cli.js mcp`
 **Verification:** Command returned JSON array with 6 simulator objects
 **Validation Summary:** Successfully discovered 6 available simulators with UUIDs, names, and boot status
 **Timestamp:** 2025-01-29 14:30:15
@@ -87,8 +87,8 @@ After testing `list_sims` tool, update the report:
 ### Pre-Testing Setup
 - Always start by building the project: `npm run build`
 - Verify Reloaderoo is available: `npx reloaderoo@latest --help`
-- Check server connectivity: `npx reloaderoo@latest inspect ping -- node build/index.js`
-- Get server information: `npx reloaderoo@latest inspect server-info -- node build/index.js`
+- Check server connectivity: `npx reloaderoo@latest inspect ping -- node build/cli.js mcp`
+- Get server information: `npx reloaderoo@latest inspect server-info -- node build/cli.js mcp`
 
 ### Systematic Testing Workflow
 1. **Create Initial Report**: Generate test report with all checkboxes unchecked
@@ -108,15 +108,15 @@ After testing `list_sims` tool, update the report:
 
 ### Tool Testing Process
 For each tool:
-1. Execute test with `npx reloaderoo@latest inspect call-tool <tool_name> --params '<json>' -- node build/index.js`
+1. Execute test with `npx reloaderoo@latest inspect call-tool <tool_name> --params '<json>' -- node build/cli.js mcp`
 2. Verify response format and content
 3. **IMMEDIATELY** update test report with result
 4. Check the box and add detailed verification summary
 5. Move to next tool
 
 ### Resource Testing Process
 For each resource:
-1. Execute test with `npx reloaderoo@latest inspect read-resource "<uri>" -- node build/index.js`
+1. Execute test with `npx reloaderoo@latest inspect read-resource "<uri>" -- node build/cli.js mcp`
 2. Verify resource accessibility and content format
 3. **IMMEDIATELY** update test report with result
 4. Check the box and add detailed verification summary

.claude/commands/rp-build-cli.md

@@ -1,7 +1,7 @@
 ---
 description: Build with rp-cli context builder → chat → implement
 repoprompt_managed: true
-repoprompt_commands_version: 5
+repoprompt_skills_version: 6
 repoprompt_variant: cli
 ---
 
@@ -40,9 +40,12 @@ rp-cli -e 'select set src/ && context'
 
 Use `rp-cli -e 'describe <tool>'` for help on a specific tool, or `rp-cli --help` for CLI usage.
 
+**⚠️ TIMEOUT WARNING:** The `builder` and `chat` commands can take several minutes to complete. When invoking rp-cli, **set your command timeout to at least 2700 seconds (45 minutes)** to avoid premature termination.
+
 ---
 ## The Workflow
 
+0. **Verify workspace** – Confirm the target codebase is loaded
 1. **Quick scan** – Understand how the task relates to the codebase
 2. **Context builder** – Call `builder` with a clear prompt to get deep context + an architectural plan
 3. **Refine with chat** – Use `chat` to clarify the plan if needed
@@ -53,26 +56,51 @@ Use `rp-cli -e 'describe <tool>'` for help on a specific tool, or `rp-cli --help
 ## CRITICAL REQUIREMENT
 
 ⚠️ **DO NOT START IMPLEMENTATION** until you have:
-1. Completed Phase 1 (Quick Scan)
-2. **Called `builder`** and received its plan
+1. Completed Phase 0 (Workspace Verification)
+2. Completed Phase 1 (Quick Scan)
+3. **Called `builder`** and received its plan
 
 Skipping `builder` results in shallow implementations that miss architectural patterns, related code, and edge cases. The quick scan alone is NOT sufficient for implementation.
 
 ---
 
+## Phase 0: Workspace Verification (REQUIRED)
+
+Before any exploration, confirm the target codebase is loaded:
+
+```bash
+# First, list available windows to find the right one
+rp-cli -e 'windows'
+
+# Then check roots in a specific window (REQUIRED - CLI cannot auto-bind)
+rp-cli -w <window_id> -e 'tree --type roots'
+```
+
+**Check the output:**
+- If your target root appears in a window → note the window ID and proceed to Phase 1
+- If not → the codebase isn't loaded in any window
+
+**CLI Window Routing (CRITICAL):**
+- CLI invocations are stateless—you MUST pass `-w <window_id>` to target the correct window
+- Use `rp-cli -e 'windows'` to list all open windows and their workspaces
+- Always include `-w <window_id>` in ALL subsequent commands
+- Without `-w`, commands may target the wrong workspace
+
+---
+
 ## Phase 1: Quick Scan (LIMITED - 2-3 tool calls max)
 
 ⚠️ **This phase is intentionally brief.** Do NOT do extensive exploration here—that's what `builder` is for.
 
 Start by getting a lay of the land with the file tree:
 ```bash
-rp-cli -e 'tree'
+rp-cli -w <window_id> -e 'tree'
 ```
 
 Then use targeted searches to understand how the task maps to the codebase:
 ```bash
-rp-cli -e 'search "<key term from task>"'
-rp-cli -e 'structure RootName/likely/relevant/area/'
+rp-cli -w <window_id> -e 'search "<key term from task>"'
+rp-cli -w <window_id> -e 'structure RootName/likely/relevant/area/'
 ```
 
 Use what you learn to **reformulate the user's prompt** with added clarity—reference specific modules, patterns, or terminology from the codebase.
@@ -86,7 +114,7 @@ Use what you learn to **reformulate the user's prompt** with added clarity—ref
 Call `builder` with your informed prompt. Use `response_type: "plan"` to get an actionable architectural plan.
 
 ```bash
-rp-cli -e 'builder "<reformulated prompt with codebase context>" --response-type plan'
+rp-cli -w <window_id> -e 'builder "<reformulated prompt with codebase context>" --response-type plan'
 ```
 
 **What you get back:**
@@ -139,21 +167,21 @@ Implement the plan directly. **Do not use `chat` with `mode:"edit"`** – you im
 **Primary tools:**
 ```bash
 # Modify existing files (search/replace) - JSON format required
-rp-cli -e 'call apply_edits {"path":"Root/File.swift","search":"old","replace":"new"}'
+rp-cli -w <window_id> -e 'call apply_edits {"path":"Root/File.swift","search":"old","replace":"new"}'
 
 # Multiline edits
-rp-cli -e 'call apply_edits {"path":"Root/File.swift","search":"old\ntext","replace":"new\ntext"}'
+rp-cli -w <window_id> -e 'call apply_edits {"path":"Root/File.swift","search":"old\ntext","replace":"new\ntext"}'
 
 # Create new files
-rp-cli -e 'file create Root/NewFile.swift "content..."'
+rp-cli -w <window_id> -e 'file create Root/NewFile.swift "content..."'
 
 # Read specific sections during implementation
-rp-cli -e 'read Root/File.swift --start-line 50 --limit 30'
+rp-cli -w <window_id> -e 'read Root/File.swift --start-line 50 --limit 30'
 ```
 
 **Ask the chat when stuck:**
 ```bash
-rp-cli -t '<tab_id>' -e 'chat "I'\''m implementing X but unsure about Y. What pattern should I follow?" --mode chat'
+rp-cli -w <window_id> -t '<tab_id>' -e 'chat "I'\''m implementing X but unsure about Y. What pattern should I follow?" --mode chat'
 ```
 
 ---
@@ -169,13 +197,13 @@ rp-cli -t '<tab_id>' -e 'chat "I'\''m implementing X but unsure about Y. What pa
 
 ```bash
 # Check current selection and tokens
-rp-cli -e 'select get'
+rp-cli -w <window_id> -e 'select get'
 
 # Add a file if needed
-rp-cli -e 'select add Root/path/to/file.swift'
+rp-cli -w <window_id> -e 'select add Root/path/to/file.swift'
 
 # Add a slice of a large file
-rp-cli -e 'select add Root/large/file.swift:100-200'
+rp-cli -w <window_id> -e 'select add Root/large/file.swift:100-200'
 ```
 
 **Chat sees only the selection:** If you need the chat's insight on a file, it must be selected first.
@@ -193,6 +221,7 @@ rp-cli -e 'select add Root/large/file.swift:100-200'
 - 🚫 **CRITICAL:** Doing extensive exploration (5+ tool calls) before calling `builder` – the quick scan should be 2-3 calls max
 - 🚫 Reading full file contents during Phase 1 – save that for after `builder` builds context
 - 🚫 Convincing yourself you understand enough to skip `builder` – you don't
+- 🚫 **CLI:** Forgetting to pass `-w <window_id>` – CLI invocations are stateless and require explicit window targeting
 
 ---
 

.claude/commands/rp-investigate-cli.md

@@ -1,7 +1,7 @@
 ---
 description: Deep codebase investigation and architecture research with rp-cli commands
 repoprompt_managed: true
-repoprompt_commands_version: 5
+repoprompt_skills_version: 6
 repoprompt_variant: cli
 ---
 
@@ -40,6 +40,8 @@ rp-cli -e 'select set src/ && context'
 
 Use `rp-cli -e 'describe <tool>'` for help on a specific tool, or `rp-cli --help` for CLI usage.
 
+**⚠️ TIMEOUT WARNING:** The `builder` and `chat` commands can take several minutes to complete. When invoking rp-cli, **set your command timeout to at least 2700 seconds (45 minutes)** to avoid premature termination.
+
 ---
 ## Investigation Protocol
 
@@ -49,6 +51,27 @@ Use `rp-cli -e 'describe <tool>'` for help on a specific tool, or `rp-cli --help
 3. **Question everything** - if something seems off, investigate it
 4. **Use `builder` aggressively** - it's designed for deep exploration
 
+### Phase 0: Workspace Verification (REQUIRED)
+
+Before any investigation, confirm the target codebase is loaded:
+
+```bash
+# First, list available windows to find the right one
+rp-cli -e 'windows'
+
+# Then check roots in a specific window (REQUIRED - CLI cannot auto-bind)
+rp-cli -w <window_id> -e 'tree --type roots'
+```
+
+**Check the output:**
+- If your target root appears in a window → note the window ID and proceed to Phase 1
+- If not → the codebase isn't loaded in any window
+
+**CLI Window Routing (CRITICAL):**
+- CLI invocations are stateless—you MUST pass `-w <window_id>` to target the correct window
+- Use `rp-cli -e 'windows'` to list all open windows and their workspaces
+- Always include `-w <window_id>` in ALL subsequent commands
+
 ### Phase 1: Initial Assessment
 
 1. Read any provided files/reports (traces, logs, error reports)
@@ -62,7 +85,7 @@ Use `rp-cli -e 'describe <tool>'` for help on a specific tool, or `rp-cli --help
 Use `builder` with detailed instructions:
 
 ```bash
-rp-cli -e 'builder "Investigate: <specific area>
+rp-cli -w <window_id> -e 'builder "Investigate: <specific area>
 
 Symptoms observed:
 - <symptom 1>
@@ -82,10 +105,10 @@ Areas to explore:
 After `builder` returns, continue with targeted questions:
 
 ```bash
-rp-cli -t '<tab_id>' -e 'chat "<specific follow-up based on findings>" --mode plan'
+rp-cli -w <window_id> -t '<tab_id>' -e 'chat "<specific follow-up based on findings>" --mode plan'
 ```
 
-> Pass `-t <tab_id>` to target the same tab across separate CLI invocations.
+> Pass `-w <window_id>` to target the correct window and `-t <tab_id>` to target the same tab across separate CLI invocations.
 
 ### Phase 4: Evidence Gathering
 
@@ -156,12 +179,14 @@ Create a findings report as you investigate:
 ## Anti-patterns to Avoid
 
 - 🚫 **CRITICAL:** Skipping `builder` and attempting to investigate by reading files manually – you'll miss critical context
+- 🚫 Skipping Phase 0 (Workspace Verification) – you must confirm the target codebase is loaded first
 - 🚫 Doing extensive exploration (5+ tool calls) before calling `builder` – initial assessment should be brief
 - 🚫 Drawing conclusions before `builder` has built proper context
 - 🚫 Reading many full files during Phase 1 – save deep reading for after `builder`
 - 🚫 Assuming you understand the issue without systematic exploration via `builder`
 - 🚫 Using only chat follow-ups without an initial `builder` call
+- 🚫 **CLI:** Forgetting to pass `-w <window_id>` – CLI invocations are stateless and require explicit window targeting
 
 ---
 
-Now begin the investigation. Read any provided context, then **immediately** use `builder` to start systematic exploration. Do not attempt manual exploration first.
+Now begin the investigation. First run `rp-cli -e 'windows'` to find the correct window, then Read any provided context, then **immediately** use `builder` to start systematic exploration. Do not attempt manual exploration first.

.claude/commands/rp-oracle-export-cli.md

@@ -1,7 +1,7 @@
 ---
 description: Export context for oracle consultation using rp-cli
 repoprompt_managed: true
-repoprompt_commands_version: 5
+repoprompt_skills_version: 6
 repoprompt_variant: cli
 ---
 
@@ -22,10 +22,31 @@ You don't need to specify which files to include—just describe what you need h
 
 ## Workflow
 
+### 0. Workspace Verification (REQUIRED)
+
+Before building context, confirm the target codebase is loaded:
+
+```bash
+# First, list available windows to find the right one
+rp-cli -e 'windows'
+
+# Then check roots in a specific window (REQUIRED - CLI cannot auto-bind)
+rp-cli -w <window_id> -e 'tree --type roots'
+```
+
+**Check the output:**
+- If your target root appears in a window → note the window ID and proceed
+- If not → the codebase isn't loaded in any window
+
+**CLI Window Routing (CRITICAL):**
+- CLI invocations are stateless—you MUST pass `-w <window_id>` to target the correct window
+- Use `rp-cli -e 'windows'` to list all open windows and their workspaces
+- Always include `-w <window_id>` in ALL subsequent commands
+
 ### 1. Build Context
 
 ```bash
-rp-cli -e 'builder "<the task/question above>" --response-type clarify'
+rp-cli -w <window_id> -e 'builder "<the task/question above>" --response-type clarify'
 ```
 
 Wait for context_builder to complete. It will explore the codebase and build optimal context.
@@ -35,7 +56,7 @@ Wait for context_builder to complete. It will explore the codebase and build opt
 Confirm the export path with the user (default: `~/Downloads/oracle-prompt.md`), then export:
 
 ```bash
-rp-cli -e 'prompt export "<confirmed path>"'
+rp-cli -w <window_id> -e 'prompt export "<confirmed path>"'
 ```
 
 Report the export path and token count to the user.