Feat: Ollama download progress tracking with new apps structure

[rayBlock]

Summary

Integrated Ollama real-time download progress tracking feature with the new apps directory structure (v2.7.2 restructure).

What's Included

• Real-time Progress Tracking: NDJSON parsing from Ollama API with speed calculation and time estimation
• UI Component: Enhanced OllamaModelSelector with animated progress bars
• IPC Integration: Streaming progress events from main process to renderer
• Comprehensive Tests: 394 lines of test coverage for parser and component
• Full Documentation: JSDoc docstrings for all new methods and types

Changes

Frontend:

• OllamaModelSelector.tsx - Download progress UI with real-time metrics
• ndjson-parser.test.ts - NDJSON streaming parser tests
• OllamaModelSelector.progress.test.ts - Component progress tests
• memory-handlers.ts - IPC event streaming
• project-api.ts - Download progress listener interface

Backend:

• Ollama Embedder provider support
• Ollama LLM provider support
• Model detection utilities

Files Changed

• 3 commits
• 12 files modified/created
• 1,023 insertions, 294 deletions

Ready For

• ✅ Merge to develop
• ✅ Integration testing
• ✅ Next release cycle

Closes #XXX

Summary by CodeRabbit

Release Notes

• New Features

  • Added CLI batch management commands for creating, monitoring, and cleaning up batch tasks.
  • Integrated Ollama model discovery, status checking, and downloading capabilities.
  • Enhanced UI with real-time progress tracking showing download speed and estimated time remaining for model downloads.

• Tests

  • Added comprehensive tests for NDJSON parsing and progress calculation utilities.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Walkthrough

This PR introduces batch task management to the CLI backend and comprehensive Ollama model management capabilities across the frontend application. It adds batch command handlers for creating, listing, and cleaning up task specifications; new IPC handlers for Ollama status and model operations; NDJSON progress streaming for real-time download tracking; and enhanced UI components with download progress visualization.

Changes

Cohort / File(s)	Summary
Batch Task Management CLI apps/backend/cli/batch_commands.py, apps/backend/cli/main.py	Introduces three new CLI commands: handle_batch_create_command validates batch JSON, creates spec directories with metadata, and tracks progress; handle_batch_status_command scans specs and infers status from file presence; handle_batch_cleanup_command identifies and optionally removes completed specs. Integrates commands into main CLI flow with new flags --batch-create, --batch-status, --batch-cleanup, and --no-dry-run.
Ollama Model Management IPC Handlers apps/frontend/src/main/ipc-handlers/memory-handlers.ts	Adds five new TypeScript interfaces (OllamaStatus, OllamaModel, OllamaEmbeddingModel, OllamaRecommendedModel, OllamaPullResult) and registers four new IPC handlers: OLLAMA_CHECK_STATUS, OLLAMA_LIST_MODELS, OLLAMA_LIST_EMBEDDING_MODELS, and OLLAMA_PULL_MODEL. Implements NDJSON-based progress streaming from Python subprocess, emitting real-time updates via OLLAMA_PULL_PROGRESS events to the renderer.
Frontend API Layer apps/frontend/src/preload/api/project-api.ts, apps/frontend/src/shared/types/ipc.ts	Adds three new preload API methods: scanOllamaModels, downloadOllamaModel, and onDownloadProgress (progress callback listener). Extends ElectronAPI type with onDownloadProgress method signature. Implementations wire to corresponding IPC channels.
NDJSON Progress Parser apps/frontend/src/main/__tests__/ndjson-parser.test.ts	Implements parseNDJSON(chunk, bufferRef) utility for streaming JSON parsing, maintaining internal buffer state across chunks, splitting by newlines, and skipping invalid/empty lines. Comprehensive test suite covers single/multiple objects, buffer management, error handling, Ollama-like data, and streaming scenarios.
Ollama Model Selector Component & Tests apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx, apps/frontend/src/renderer/components/__tests__/OllamaModelSelector.progress.test.ts	Enhances OllamaModelSelector component with real-time download progress tracking using refs and state. Adds progress calculation utilities (calculateSpeed, formatSpeed, calculateTimeRemaining, formatTimeRemaining, calculatePercentage) with comprehensive test coverage. Updates UI to display progress bars, speed, and ETA during model downloads.
Infrastructure Mocks & Types apps/frontend/src/renderer/lib/mocks/infrastructure-mock.ts	Expands embedding models list from 1 to 3 models via mock listOllamaEmbeddingModels. Adds onDownloadProgress mock callback API that stores callback on global window for test verification and returns cleanup function.

Sequence Diagram(s)

sequenceDiagram
    participant Renderer as OllamaModelSelector<br/>(Renderer)
    participant Preload as Project API<br/>(Preload)
    participant Main as Memory Handler<br/>(Main Process)
    participant Python as ollama_model_<br/>detector.py
    participant Ollama as Ollama<br/>Service

    Renderer->>+Preload: downloadOllamaModel(baseUrl,<br/>modelName)
    Preload->>+Main: ipcRenderer.invoke<br/>('download-ollama-model')
    Main->>+Python: spawn subprocess with<br/>model pull command
    Note over Main,Python: Setup stderr listener<br/>for NDJSON progress
    Python->>+Ollama: initiate model pull
    
    loop Progress Streaming
        Python-->>Python: emit JSON status<br/>to stderr
        Main->>Main: parseNDJSON(line)
        Main->>Renderer: emit IPC_CHANNELS.<br/>OLLAMA_PULL_PROGRESS
        Renderer->>Renderer: update progress state<br/>& calculate speed/ETA
        Renderer->>Renderer: re-render with<br/>progress bar
    end
    
    Ollama-->>Python: model download complete
    Python-->>Main: exit with success
    Main->>Main: emit final success<br/>event
    Main-->>Preload: return OllamaPullResult
    Preload-->>Renderer: resolve promise
    Renderer->>Renderer: refresh installed<br/>models list
    Renderer->>Renderer: clear progress,<br/>update UI

Loading

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

The changes span multiple layers (backend CLI, main process IPC, preload API, renderer components) with heterogeneous logic patterns: batch filesystem orchestration, subprocess streaming with NDJSON parsing, real-time progress calculations, and stateful UI component updates. Reviewers must verify correct IPC channel wiring, progress calculation accuracy, buffer management in streaming scenarios, and proper lifecycle management of progress listeners across the application.

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately reflects the main change: adding Ollama download progress tracking with the new apps directory structure, which is a primary focus across multiple file changes.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[AndyMik90]

@CodeRabbit review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[AndyMik90]

@rayBlock

Quick note: the PR summary mentions Ollama download progress tracking, but the changes also include a new batch task management system (batch_commands.py with --batch-create, --batch-status, --batch-cleanup CLI commands).

Might be worth updating the description to include that feature so reviewers know to look at it. 👍

[coderabbitai]

Actionable comments posted: 6

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (2)

apps/frontend/src/main/ipc-handlers/memory-handlers.ts (2)

394-457: Duplicate IPC handler registration will crash the application.

ipcMain.handle throws an error when registering a handler for the same channel twice: "Attempted to register a second handler for '${channel}'". Lines 394-419 and 433-457 both register handlers for IPC_CHANNELS.OLLAMA_CHECK_STATUS.

Remove the first (undocumented) handler and keep the second one with JSDoc.

🔎 Proposed fix to remove duplicate handler

   // ============================================
   // Ollama Model Detection Handlers
   // ============================================

-  // Check if Ollama is running
-  ipcMain.handle(
-    IPC_CHANNELS.OLLAMA_CHECK_STATUS,
-    async (_, baseUrl?: string): Promise<IPCResult<OllamaStatus>> => {
-      try {
-        const result = await executeOllamaDetector('check-status', baseUrl);
-
-        if (!result.success) {
-          return {
-            success: false,
-            error: result.error || 'Failed to check Ollama status',
-          };
-        }
-
-        return {
-          success: true,
-          data: result.data as OllamaStatus,
-        };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to check Ollama status',
-        };
-      }
-    }
-  );
-
    // ============================================
    // Ollama Model Discovery & Management
    // ============================================

---

575-600: Path array is inconsistent and timeout option is invalid for spawn().

Two issues:

1. Inconsistent paths: executeOllamaDetector (lines 113-120) searches both apps/backend and legacy auto-claude paths, but this handler only checks legacy paths. The script won't be found in the new directory structure.

2. Invalid timeout: Node.js spawn() does not support the timeout option—only execFile() and exec() do. The 10-minute timeout will be silently ignored, leaving large downloads without timeout protection.

🔎 Proposed fix

         // Find the ollama_model_detector.py script
         const possiblePaths = [
+          // Development paths
+          path.resolve(__dirname, '..', '..', '..', '..', 'backend', 'ollama_model_detector.py'),
+          path.resolve(process.cwd(), 'apps', 'backend', 'ollama_model_detector.py'),
+          // Legacy paths (for backwards compatibility)
           path.resolve(__dirname, '..', '..', '..', 'auto-claude', 'ollama_model_detector.py'),
           path.resolve(process.cwd(), 'auto-claude', 'ollama_model_detector.py'),
           path.resolve(process.cwd(), '..', 'auto-claude', 'ollama_model_detector.py'),
         ];

For the timeout, implement manual timeout handling similar to executeOllamaDetector:

         return new Promise((resolve) => {
+          let resolved = false;
           const proc = spawn(pythonExe, args, {
             stdio: ['ignore', 'pipe', 'pipe'],
-            timeout: 600000, // 10 minute timeout for large models
           });

+          const timeoutId = setTimeout(() => {
+            if (!resolved) {
+              resolved = true;
+              proc.kill();
+              resolve({ success: false, error: 'Timeout: model download took longer than 10 minutes' });
+            }
+          }, 600000);

           // ... existing code ...

           proc.on('close', (code) => {
+            if (resolved) return;
+            resolved = true;
+            clearTimeout(timeoutId);
             // ... existing handler code ...
           });

           proc.on('error', (err) => {
+            if (resolved) return;
+            resolved = true;
+            clearTimeout(timeoutId);
             resolve({ success: false, error: err.message });
           });
         });

🧹 Nitpick comments (9)

apps/frontend/src/main/__tests__/ndjson-parser.test.ts (2)

34-54: Consider importing the actual parser to avoid drift.

The parseNDJSON function here is a duplicate of the production implementation in memory-handlers.ts. If the production code changes, these tests won't catch regressions. Consider either:

1. Exporting the parser from memory-handlers.ts and importing it here
2. Moving the parser to a shared utility module that both files can import

This ensures tests validate the actual production code.

---

207-221: Type safety: accessing property not defined in ProgressData interface.

Line 219 accesses results[0].other, but other is not defined in the ProgressData interface. While this works at runtime due to JSON.parse returning any, it bypasses TypeScript's type checking. Consider extending the interface or using a type assertion to document this intentional behavior.

🔎 Suggested fix

 interface ProgressData {
   status?: string;    // Current operation (e.g., 'downloading', 'extracting', 'verifying')
   completed?: number; // Bytes downloaded so far
   total?: number;     // Total bytes to download
+  [key: string]: unknown; // Allow additional properties for testing
 }

apps/backend/cli/main.py (1)

315-326: Consider using return values for exit codes.

The batch command handlers return bool to indicate success/failure, but these return values are ignored. For CLI consistency, consider propagating failures via sys.exit(1).

🔎 Suggested improvement

     # Handle batch commands
     if args.batch_create:
-        handle_batch_create_command(args.batch_create, str(project_dir))
-        return
+        if not handle_batch_create_command(args.batch_create, str(project_dir)):
+            sys.exit(1)
+        return

     if args.batch_status:
-        handle_batch_status_command(str(project_dir))
-        return
+        if not handle_batch_status_command(str(project_dir)):
+            sys.exit(1)
+        return

     if args.batch_cleanup:
-        handle_batch_cleanup_command(str(project_dir), dry_run=not args.no_dry_run)
-        return
+        if not handle_batch_cleanup_command(str(project_dir), dry_run=not args.no_dry_run):
+            sys.exit(1)
+        return

apps/frontend/src/renderer/components/__tests__/OllamaModelSelector.progress.test.ts (2)

8-11: Unused interface: ProgressTracking is defined but never used.

This interface appears to be leftover from development. Consider removing it to keep the test file clean.

🔎 Suggested fix

-interface ProgressTracking {
-  lastCompleted: number;
-  lastUpdate: number;
-}
-

---

13-92: Extract utilities to a shared module for reuse and single source of truth.

These calculation functions duplicate logic from OllamaModelSelector.tsx. Consider extracting them to a shared utility file (e.g., progress-utils.ts) that both the component and tests can import. This ensures:

1. Tests validate actual production code
2. No drift between implementations
3. Utilities are reusable elsewhere

apps/frontend/src/renderer/lib/mocks/infrastructure-mock.ts (1)

83-118: Indentation inconsistency: 3 spaces instead of 2.

Lines 83-118 use 3-space indentation while the rest of the file uses 2-space indentation. This should be corrected for consistency.

🔎 Suggested fix

-   listOllamaEmbeddingModels: async () => ({
-     success: true,
-     data: {
-       embedding_models: [
-         { name: 'embeddinggemma', embedding_dim: 768, description: "Google's lightweight embedding model (Recommended)", size_bytes: 650000000, size_gb: 0.621 },
-         { name: 'nomic-embed-text', embedding_dim: 768, description: 'Popular general-purpose embeddings', size_bytes: 287000000, size_gb: 0.274 },
-         { name: 'mxbai-embed-large', embedding_dim: 1024, description: 'MixedBread AI large embeddings', size_bytes: 701000000, size_gb: 0.670 },
-       ],
-       count: 3
-     }
-   }),
+  listOllamaEmbeddingModels: async () => ({
+    success: true,
+    data: {
+      embedding_models: [
+        { name: 'embeddinggemma', embedding_dim: 768, description: "Google's lightweight embedding model (Recommended)", size_bytes: 650000000, size_gb: 0.621 },
+        { name: 'nomic-embed-text', embedding_dim: 768, description: 'Popular general-purpose embeddings', size_bytes: 287000000, size_gb: 0.274 },
+        { name: 'mxbai-embed-large', embedding_dim: 1024, description: 'MixedBread AI large embeddings', size_bytes: 701000000, size_gb: 0.670 },
+      ],
+      count: 3
+    }
+  }),

Apply the same 2-space indentation fix to pullOllamaModel and onDownloadProgress as well.

apps/backend/cli/batch_commands.py (1)

10-10: Unused import: Optional is imported but never used.

🔎 Suggested fix

-from typing import Optional

apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx (2)

181-266: Indentation inconsistency: mixed 2-space and 3-space indentation.

Several blocks in this file use 3-space indentation (lines 181-266, 268-304, 348-456) while the rest uses 2-space. This should be standardized for maintainability.

---

209-235: Progress calculation logic duplicated from test utilities.

The speed formatting and time remaining logic here duplicates what's in OllamaModelSelector.progress.test.ts. Consider extracting to a shared utility module for single source of truth and easier testing.

📜 Review details

Configuration used: defaults

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 0adadda and fed2cdd.

📒 Files selected for processing (9)

• apps/backend/cli/batch_commands.py
• apps/backend/cli/main.py
• apps/frontend/src/main/__tests__/ndjson-parser.test.ts
• apps/frontend/src/main/ipc-handlers/memory-handlers.ts
• apps/frontend/src/preload/api/project-api.ts
• apps/frontend/src/renderer/components/__tests__/OllamaModelSelector.progress.test.ts
• apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx
• apps/frontend/src/renderer/lib/mocks/infrastructure-mock.ts
• apps/frontend/src/shared/types/ipc.ts

🧰 Additional context used

🧬 Code graph analysis (2)

apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx (3)

.design-system/src/lib/utils.ts (1)

• cn (4-6)

.design-system/src/lib/icons.ts (1)

• Check (21-21)

.design-system/src/components/Button.tsx (1)

• Button (10-44)

apps/backend/cli/main.py (1)

apps/backend/cli/batch_commands.py (3)

• handle_batch_cleanup_command (169-210)
• handle_batch_create_command (14-103)
• handle_batch_status_command (106-166)

🪛 Biome (2.1.2)

apps/frontend/src/preload/api/project-api.ts

[error] 236-236: Expected an expression but instead found '<<'.

Expected an expression here.

(parse)

---

[error] 236-236: Expected an expression but instead found '<<'.

Expected an expression here.

(parse)

---

[error] 236-236: expected > but instead found :

Remove :

(parse)

---

[error] 246-249: Expected a statement but instead found '>> =>
ipcRenderer.invoke('scan-ollama-models', baseUrl),

downloadOllamaModel: (baseUrl: string, modelName: string): Promise<IPCResult<'.

Expected a statement here.

(parse)

---

[error] 249-252: Expected a statement but instead found '>> =>
ipcRenderer.invoke('download-ollama-model', baseUrl, modelName),

onDownloadProgress: (callback: (data:'.

Expected a statement here.

(parse)

---

[error] 258-258: Expected a statement but instead found ') => void) =>'.

Expected a statement here.

(parse)

---

[error] 261-261: Illegal return statement outside of a function

(parse)

---

[error] 262-264: Expected a statement but instead found ',

795c470 (feat(ollama): add real-time download progress tracking'.

Expected a statement here.

(parse)

---

[error] 264-264: numbers cannot be followed by identifiers directly after

an identifier cannot appear here

(parse)

---

[error] 264-264: expected ( but instead found model

Remove model

(parse)

---

[error] 264-264: expected ; but instead found downloads

Remove downloads

(parse)

---

[error] 264-264: expected ; but instead found )

Remove )

(parse)

---

[error] 264-264: Expected a statement but instead found ':'.

Expected a statement here.

(parse)

🔇 Additional comments (9)

apps/frontend/src/shared/types/ipc.ts (1)

581-591: LGTM!

The onDownloadProgress API follows the established pattern for event listeners in this interface, returning a cleanup function for proper resource management. The data shape is well-defined with all necessary progress metrics.

apps/backend/cli/main.py (1)

245-267: LGTM! Well-structured batch command arguments.

The CLI arguments are clearly documented and follow the existing patterns in the codebase. The --no-dry-run flag correctly inverts for safety-first defaults.

apps/frontend/src/renderer/components/__tests__/OllamaModelSelector.progress.test.ts (1)

94-195: Excellent test coverage for progress calculations.

The tests comprehensively cover edge cases including zero/negative values, unit scaling thresholds, and realistic download scenarios. The real-world scenario tests at the end provide valuable integration-level validation.

apps/frontend/src/renderer/lib/mocks/infrastructure-mock.ts (1)

104-118: Mock implementation is functional and follows the expected API contract.

The onDownloadProgress mock correctly stores the callback for test verification and returns a cleanup function. This pattern enables tests to verify progress handling behavior.

apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx (1)

431-452: Well-implemented progress bar UI with proper bounds clamping.

The progress bar correctly clamps percentage between 0-100 and displays speed/time remaining metrics. The gradient styling provides good visual feedback during downloads.

apps/frontend/src/main/ipc-handlers/memory-handlers.ts (4)

28-85: LGTM!

The interface definitions are well-documented with clear JSDoc comments explaining each field's purpose. The type structure appropriately captures Ollama model metadata.

---

87-102: LGTM!

Comprehensive JSDoc documentation covering the function's purpose, supported commands, parameters, and return type.

---

610-646: LGTM!

The NDJSON streaming parser correctly handles:

• Buffering incomplete lines across chunks (line 619)
• Safe JSON parsing with silent skip for non-JSON lines
• Division-by-zero protection in percentage calculation (line 628)
• Progress event emission with all required fields

---

467-541: LGTM!

The model listing handlers correctly delegate to executeOllamaDetector and extract the relevant data from the response. Error handling is consistent with the rest of the codebase.

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Potential crash when parsing spec IDs.

Line 51 assumes spec names starting with a digit contain a hyphen, but s.split("-")[0] will return the whole string if no hyphen exists, and int() will raise ValueError for non-numeric strings like "1abc".

🔎 Suggested fix

     # Find next spec ID
     existing_specs = [d.name for d in specs_dir.iterdir() if d.is_dir()]
-    next_id = max([int(s.split("-")[0]) for s in existing_specs if s[0].isdigit()] or [0]) + 1
+    def extract_spec_id(name: str) -> int | None:
+        """Extract numeric ID from spec name like '001-feature'."""
+        parts = name.split("-", 1)
+        try:
+            return int(parts[0])
+        except ValueError:
+            return None
+    
+    spec_ids = [id for name in existing_specs if (id := extract_spec_id(name)) is not None]
+    next_id = max(spec_ids or [0]) + 1

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	# Find next spec ID
	existing_specs = [d.name for d in specs_dir.iterdir() if d.is_dir()]
	next_id = max([int(s.split("-")[0]) for s in existing_specs if s[0].isdigit()] or [0]) + 1
	# Find next spec ID
	existing_specs = [d.name for d in specs_dir.iterdir() if d.is_dir()]
	def extract_spec_id(name: str) -> int | None:
	"""Extract numeric ID from spec name like '001-feature'."""
	parts = name.split("-", 1)
	try:
	return int(parts[0])
	except ValueError:
	return None
	spec_ids = [id for name in existing_specs if (id := extract_spec_id(name)) is not None]
	next_id = max(spec_ids or [0]) + 1

🤖 Prompt for AI Agents

In apps/backend/cli/batch_commands.py around lines 49-51, the current code
assumes spec folder names always contain a hyphen and a numeric prefix which can
raise ValueError for names like "1abc"; change the logic to only consider names
with a numeric prefix by either filtering with a regex that matches leading
digits followed by a hyphen (e.g. r'^\d+-') or extracting the leading digit
sequence with a regex (r'^(\d+)') and converting that group to int inside a safe
path (skip non-matching names); compute next_id as max(parsed_ints or [0]) + 1
so non-numeric names cannot cause a crash.

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Slug sanitization missing: potential filesystem issues.

task_slug only replaces spaces with hyphens and truncates, but doesn't handle special characters that could cause filesystem issues (e.g., /, \, :, *, ?, ", <, >, |).

🔎 Suggested fix

+import re
+
+def sanitize_slug(title: str) -> str:
+    """Create filesystem-safe slug from title."""
+    slug = title.lower().replace(" ", "-")
+    slug = re.sub(r'[^\w\-]', '', slug)  # Remove non-alphanumeric except hyphen
+    return slug[:50]
+
 # In handle_batch_create_command:
         task_title = task.get("title", f"Task {idx}")
-        task_slug = task_title.lower().replace(" ", "-")[:50]
+        task_slug = sanitize_slug(task_title)

Committable suggestion skipped: line range outside the PR's diff.

🤖 Prompt for AI Agents

In apps/backend/cli/batch_commands.py around lines 57-58, the current slug
generation only lowercases, replaces spaces and truncates, which leaves unsafe
filesystem characters; update the logic to (1) normalize the title (NFKD) and
remove/replace any characters not allowed in filenames by replacing any run of
characters other than ASCII letters, numbers and hyphens with a single hyphen
(use a regex), (2) collapse multiple hyphens to one and strip leading/trailing
hyphens, (3) ensure the result is truncated to 50 chars and fall back to a safe
default like "task-{idx}" if empty, or use a well-tested slugify utility if
available instead of hand-rolling.

[coderabbitai]

⚠️ Potential issue | 🟠 Major

Status determination order is incorrect.

The status checks are in the wrong order. A spec with qa_report.md will incorrectly show as spec_created if spec.md also exists (which it should for QA-approved specs). The check for qa_report.md should come first.

🔎 Suggested fix

         # Determine status
-        if (spec_dir / "spec.md").exists():
-            status = "spec_created"
-        elif (spec_dir / "implementation_plan.json").exists():
-            status = "building"
-        elif (spec_dir / "qa_report.md").exists():
+        if (spec_dir / "qa_report.md").exists():
             status = "qa_approved"
+        elif (spec_dir / "implementation_plan.json").exists():
+            status = "building"
+        elif (spec_dir / "spec.md").exists():
+            status = "spec_created"
         else:
             status = "pending_spec"

🤖 Prompt for AI Agents

In apps/backend/cli/batch_commands.py around lines 146 to 154, the status checks
are ordered incorrectly causing QA-approved specs to be reported as
spec_created; reorder the checks so that qa_report.md is checked first (set
status to "qa_approved"), then implementation_plan.json (set "building"), then
spec.md (set "spec_created"), and finally default to "pending_spec" — keep using
the same Path.exists calls but change the if/elif sequence accordingly.

[coderabbitai]

⚠️ Potential issue | 🟠 Major

handle_batch_cleanup_command only implements dry-run mode.

The function accepts dry_run parameter but only implements the dry-run path (lines 199-208). When dry_run=False, the function prints the count but doesn't actually delete anything.

🔎 Suggested fix

     if dry_run:
         print()
         print("Would remove:")
         for spec_name in completed:
             print(f"  - {spec_name}")
             wt_path = worktrees_dir / spec_name
             if wt_path.exists():
                 print(f"    └─ .worktrees/{spec_name}/")
         print()
         print("Run with --no-dry-run to actually delete")
+    else:
+        import shutil
+        for spec_name in completed:
+            spec_path = specs_dir / spec_name
+            wt_path = worktrees_dir / spec_name
+            
+            if wt_path.exists():
+                shutil.rmtree(wt_path)
+                print_status(f"Removed worktree: .worktrees/{spec_name}/", "success")
+            
+            shutil.rmtree(spec_path)
+            print_status(f"Removed spec: {spec_name}", "success")
+        
+        print()
+        print_status(f"Cleaned up {len(completed)} spec(s)", "success")

     return True

🤖 Prompt for AI Agents

In apps/backend/cli/batch_commands.py around lines 169 to 210, the function
handle_batch_cleanup_command only implements dry-run output and never performs
deletions when dry_run is False; update it so when dry_run is False it actually
deletes each completed spec directory and its corresponding worktree (e.g.,
remove specs_dir/spec_name and worktrees_dir/spec_name), using safe removal
(shutil.rmtree) and checking existence before deleting, log each deletion with
print_status or similar, catch and log exceptions per-item and return False on
any failure, and return True on success.

[coderabbitai]

⚠️ Potential issue | 🔴 Critical

Unresolved merge conflict markers will break the build.

The file contains Git merge conflict markers that must be resolved before merging. The static analysis errors confirm the code cannot be parsed.

Based on the PR objectives and interface declarations above (lines 77-93), the intended resolution appears to keep the Ollama model management implementations. Remove the conflict markers and retain the implementation:

🔎 Proposed fix to resolve merge conflict

-<<<<<<< HEAD:apps/frontend/src/preload/api/project-api.ts
-=======
   // Ollama Model Management
   scanOllamaModels: (baseUrl: string): Promise<IPCResult<{
     models: Array<{
       name: string;
       size: number;
       modified_at: string;
       digest: string;
     }>;
   }>> =>
     ipcRenderer.invoke('scan-ollama-models', baseUrl),

   downloadOllamaModel: (baseUrl: string, modelName: string): Promise<IPCResult<{ message: string }>> =>
     ipcRenderer.invoke('download-ollama-model', baseUrl, modelName),

   onDownloadProgress: (callback: (data: {
     modelName: string;
     status: string;
     completed: number;
     total: number;
     percentage: number;
   }) => void) => {
     const listener = (_: any, data: any) => callback(data);
     ipcRenderer.on(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
     return () => ipcRenderer.off(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
   },

->>>>>>> 795c470 (feat(ollama): add real-time download progress tracking for model downloads):auto-claude-ui/src/preload/api/project-api.ts
   // Git Operations

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	<<<<<<< HEAD:apps/frontend/src/preload/api/project-api.ts
	=======
	// Ollama Model Management
	scanOllamaModels: (baseUrl: string): Promise<IPCResult<{
	models: Array<{
	name: string;
	size: number;
	modified_at: string;
	digest: string;
	}>;
	}>> =>
	ipcRenderer.invoke('scan-ollama-models', baseUrl),
	downloadOllamaModel: (baseUrl: string, modelName: string): Promise<IPCResult<{ message: string }>> =>
	ipcRenderer.invoke('download-ollama-model', baseUrl, modelName),
	onDownloadProgress: (callback: (data: {
	modelName: string;
	status: string;
	completed: number;
	total: number;
	percentage: number;
	}) => void) => {
	const listener = (_: any, data: any) => callback(data);
	ipcRenderer.on(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
	return () => ipcRenderer.off(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
	},
	>>>>>>> 795c470 (feat(ollama): add real-time download progress tracking for model downloads):auto-claude-ui/src/preload/api/project-api.ts
	// Ollama Model Management
	scanOllamaModels: (baseUrl: string): Promise<IPCResult<{
	models: Array<{
	name: string;
	size: number;
	modified_at: string;
	digest: string;
	}>;
	}>> =>
	ipcRenderer.invoke('scan-ollama-models', baseUrl),
	downloadOllamaModel: (baseUrl: string, modelName: string): Promise<IPCResult<{ message: string }>> =>
	ipcRenderer.invoke('download-ollama-model', baseUrl, modelName),
	onDownloadProgress: (callback: (data: {
	modelName: string;
	status: string;
	completed: number;
	total: number;
	percentage: number;
	}) => void) => {
	const listener = (_: any, data: any) => callback(data);
	ipcRenderer.on(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
	return () => ipcRenderer.off(IPC_CHANNELS.OLLAMA_PULL_PROGRESS, listener);
	},

🧰 Tools

🪛 Biome (2.1.2)

[error] 236-236: Expected an expression but instead found '<<'.

Expected an expression here.

(parse)

---

[error] 236-236: Expected an expression but instead found '<<'.

Expected an expression here.

(parse)

---

[error] 236-236: expected > but instead found :

Remove :

(parse)

---

[error] 246-249: Expected a statement but instead found '>> =>
ipcRenderer.invoke('scan-ollama-models', baseUrl),

downloadOllamaModel: (baseUrl: string, modelName: string): Promise<IPCResult<'.

Expected a statement here.

(parse)

---

[error] 249-252: Expected a statement but instead found '>> =>
ipcRenderer.invoke('download-ollama-model', baseUrl, modelName),

onDownloadProgress: (callback: (data:'.

Expected a statement here.

(parse)

---

[error] 258-258: Expected a statement but instead found ') => void) =>'.

Expected a statement here.

(parse)

---

[error] 261-261: Illegal return statement outside of a function

(parse)

---

[error] 262-264: Expected a statement but instead found ',

795c470 (feat(ollama): add real-time download progress tracking'.

Expected a statement here.

(parse)

---

[error] 264-264: numbers cannot be followed by identifiers directly after

an identifier cannot appear here

(parse)

---

[error] 264-264: expected ( but instead found model

Remove model

(parse)

---

[error] 264-264: expected ; but instead found downloads

Remove downloads

(parse)

---

[error] 264-264: expected ; but instead found )

Remove )

(parse)

---

[error] 264-264: Expected a statement but instead found ':'.

Expected a statement here.

(parse)

🤖 Prompt for AI Agents

In apps/frontend/src/preload/api/project-api.ts around lines 236 to 264 there
are unresolved Git conflict markers (<<<<<<<, =======, >>>>>>>) that break
parsing; remove the conflict markers and retain the intended Ollama model
management implementations (scanOllamaModels, downloadOllamaModel,
onDownloadProgress) as shown in the incoming block, ensuring the function
signatures, ipcRenderer.invoke calls and IPC_CHANNELS.OLLAMA_PULL_PROGRESS
listener registration/unregistration remain intact and types match the
surrounding interface declarations.

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Missing AbortSignal when refreshing models after download.

Line 283 calls checkInstalledModels() without an AbortSignal. If the component unmounts during the refresh, this could lead to state updates on an unmounted component.

🔎 Suggested fix

+  // Store abort controller for cleanup
+  const abortControllerRef = useRef<AbortController | null>(null);
+
   const handleDownload = async (modelName: string) => {
     setIsDownloading(modelName);
     setError(null);

     try {
       const result = await window.electronAPI.pullOllamaModel(modelName);
       if (result?.success) {
         // Refresh the model list
-        await checkInstalledModels();
+        abortControllerRef.current = new AbortController();
+        await checkInstalledModels(abortControllerRef.current.signal);
       } else {
         setError(result?.error || `Failed to download ${modelName}`);
       }
     } catch (err) {
       setError(err instanceof Error ? err.message : 'Download failed');
     } finally {
       setIsDownloading(null);
+      // Clean up progress tracking for this model
+      delete downloadProgressRef.current[modelName];
+      setDownloadProgress(prev => {
+        const { [modelName]: _, ...rest } = prev;
+        return rest;
+      });
     }
   };

Committable suggestion skipped: line range outside the PR's diff.

🤖 Prompt for AI Agents

In apps/frontend/src/renderer/components/onboarding/OllamaModelSelector.tsx
around lines 275-292, the call to checkInstalledModels() after a successful
download lacks an AbortSignal and can update state after the component unmounts;
update checkInstalledModels to accept an optional AbortSignal (and honor it
inside the function), create or reuse a component-level AbortController (store
it in a ref and abort it in useEffect cleanup), and pass controller.signal into
checkInstalledModels() in handleDownload so the refresh can be cancelled if the
component unmounts.