autoforge-ai 0.1.4 → 0.1.5

package/.env.example

@@ -9,11 +9,6 @@
 # - webkit: Safari engine
 # - msedge: Microsoft Edge
 # PLAYWRIGHT_BROWSER=firefox
-#
-# PLAYWRIGHT_HEADLESS: Run browser without visible window
-# - true: Browser runs in background, saves CPU (default)
-# - false: Browser opens a visible window (useful for debugging)
-# PLAYWRIGHT_HEADLESS=true
 
 # Extra Read Paths (Optional)
 # Comma-separated list of absolute paths for read-only access to external directories.
@@ -25,56 +20,17 @@
 # Google Cloud Vertex AI Configuration (Optional)
 # To use Claude via Vertex AI on Google Cloud Platform, uncomment and set these variables.
 # Requires: gcloud CLI installed and authenticated (run: gcloud auth application-default login)
-# Note: Use @ instead of - in model names (e.g., claude-opus-4-5@20251101)
+# Note: Use @ instead of - in model names for date-suffixed models (e.g., claude-sonnet-4-5@20250929)
 #
 # CLAUDE_CODE_USE_VERTEX=1
 # CLOUD_ML_REGION=us-east5
 # ANTHROPIC_VERTEX_PROJECT_ID=your-gcp-project-id
-# ANTHROPIC_DEFAULT_OPUS_MODEL=claude-opus-4-5@20251101
+# ANTHROPIC_DEFAULT_OPUS_MODEL=claude-opus-4-6
 # ANTHROPIC_DEFAULT_SONNET_MODEL=claude-sonnet-4-5@20250929
 # ANTHROPIC_DEFAULT_HAIKU_MODEL=claude-3-5-haiku@20241022
 
 # ===================
-# Alternative API Providers
+# Alternative API Providers (GLM, Ollama, Kimi, Custom)
 # ===================
-# NOTE: These env vars are the legacy way to configure providers.
-# The recommended way is to use the Settings UI (API Provider section).
-# UI settings take precedence when api_provider != "claude".
-
-# Kimi K2.5 (Moonshot) Configuration (Optional)
-# Get an API key at: https://kimi.com
-#
-# ANTHROPIC_BASE_URL=https://api.kimi.com/coding/
-# ANTHROPIC_API_KEY=your-kimi-api-key
-# ANTHROPIC_DEFAULT_SONNET_MODEL=kimi-k2.5
-# ANTHROPIC_DEFAULT_OPUS_MODEL=kimi-k2.5
-# ANTHROPIC_DEFAULT_HAIKU_MODEL=kimi-k2.5
-
-# GLM/Alternative API Configuration (Optional)
-# To use Zhipu AI's GLM models instead of Claude, uncomment and set these variables.
-# This only affects AutoForge - your global Claude Code settings remain unchanged.
-# Get an API key at: https://z.ai/subscribe
-#
-# ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
-# ANTHROPIC_AUTH_TOKEN=your-zhipu-api-key
-# API_TIMEOUT_MS=3000000
-# ANTHROPIC_DEFAULT_SONNET_MODEL=glm-4.7
-# ANTHROPIC_DEFAULT_OPUS_MODEL=glm-4.7
-# ANTHROPIC_DEFAULT_HAIKU_MODEL=glm-4.5-air
-
-# Ollama Local Model Configuration (Optional)
-# To use local models via Ollama instead of Claude, uncomment and set these variables.
-# Requires Ollama v0.14.0+ with Anthropic API compatibility.
-# See: https://ollama.com/blog/claude
-#
-# ANTHROPIC_BASE_URL=http://localhost:11434
-# ANTHROPIC_AUTH_TOKEN=ollama
-# API_TIMEOUT_MS=3000000
-# ANTHROPIC_DEFAULT_SONNET_MODEL=qwen3-coder
-# ANTHROPIC_DEFAULT_OPUS_MODEL=qwen3-coder
-# ANTHROPIC_DEFAULT_HAIKU_MODEL=qwen3-coder
-#
-# Model recommendations:
-# - For best results, use a capable coding model like qwen3-coder or deepseek-coder-v2
-# - You can use the same model for all tiers, or different models per tier
-# - Larger models (70B+) work best for Opus tier, smaller (7B-20B) for Haiku
+# Configure alternative providers via the Settings UI (gear icon > API Provider).
+# The Settings UI is the recommended way to switch providers and models.

package/README.md

@@ -326,37 +326,13 @@ When test progress increases, the agent sends:
 }
 ```
 
-### Using GLM Models (Alternative to Claude)
+### Alternative API Providers (GLM, Ollama, Kimi, Custom)
 
-Add these variables to your `.env` file to use Zhipu AI's GLM models:
+Alternative providers are configured via the **Settings UI** (gear icon > API Provider). Select your provider, set the base URL, auth token, and model directly in the UI — no `.env` changes needed.
 
-```bash
-ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
-ANTHROPIC_AUTH_TOKEN=your-zhipu-api-key
-API_TIMEOUT_MS=3000000
-ANTHROPIC_DEFAULT_SONNET_MODEL=glm-4.7
-ANTHROPIC_DEFAULT_OPUS_MODEL=glm-4.7
-ANTHROPIC_DEFAULT_HAIKU_MODEL=glm-4.5-air
-```
-
-This routes AutoForge's API requests through Zhipu's Claude-compatible API, allowing you to use GLM-4.7 and other models. **This only affects AutoForge** - your global Claude Code settings remain unchanged.
-
-Get an API key at: https://z.ai/subscribe
-
-### Using Ollama Local Models
-
-Add these variables to your `.env` file to run agents with local models via Ollama v0.14.0+:
-
-```bash
-ANTHROPIC_BASE_URL=http://localhost:11434
-ANTHROPIC_AUTH_TOKEN=ollama
-API_TIMEOUT_MS=3000000
-ANTHROPIC_DEFAULT_SONNET_MODEL=qwen3-coder
-ANTHROPIC_DEFAULT_OPUS_MODEL=qwen3-coder
-ANTHROPIC_DEFAULT_HAIKU_MODEL=qwen3-coder
-```
+Available providers: **Claude** (default), **GLM** (Zhipu AI), **Ollama** (local models), **Kimi** (Moonshot), **Custom**
 
-See the [CLAUDE.md](CLAUDE.md) for recommended models and known limitations.
+For Ollama, install [Ollama v0.14.0+](https://ollama.com), run `ollama serve`, and pull a coding model (e.g., `ollama pull qwen3-coder`). Then select "Ollama" in the Settings UI.
 
 ### Using Vertex AI
 
@@ -366,7 +342,7 @@ Add these variables to your `.env` file to run agents via Google Cloud Vertex AI
 CLAUDE_CODE_USE_VERTEX=1
 CLOUD_ML_REGION=us-east5
 ANTHROPIC_VERTEX_PROJECT_ID=your-gcp-project-id
-ANTHROPIC_DEFAULT_OPUS_MODEL=claude-opus-4-5@20251101
+ANTHROPIC_DEFAULT_OPUS_MODEL=claude-opus-4-6
 ANTHROPIC_DEFAULT_SONNET_MODEL=claude-sonnet-4-5@20250929
 ANTHROPIC_DEFAULT_HAIKU_MODEL=claude-3-5-haiku@20241022
 ```

package/client.py

@@ -46,8 +46,9 @@ def convert_model_for_vertex(model: str) -> str:
     """
     Convert model name format for Vertex AI compatibility.
 
-    Vertex AI uses @ to separate model name from version (e.g., claude-opus-4-5@20251101)
-    while the Anthropic API uses - (e.g., claude-opus-4-5-20251101).
+    Vertex AI uses @ to separate model name from version (e.g., claude-sonnet-4-5@20250929)
+    while the Anthropic API uses - (e.g., claude-sonnet-4-5-20250929).
+    Models without a date suffix (e.g., claude-opus-4-6) pass through unchanged.
 
     Args:
         model: Model name in Anthropic format (with hyphens)
@@ -61,7 +62,7 @@ def convert_model_for_vertex(model: str) -> str:
         return model
 
     # Pattern: claude-{name}-{version}-{date} -> claude-{name}-{version}@{date}
-    # Example: claude-opus-4-5-20251101 -> claude-opus-4-5@20251101
+    # Example: claude-sonnet-4-5-20250929 -> claude-sonnet-4-5@20250929
     # The date is always 8 digits at the end
     match = re.match(r'^(claude-.+)-(\d{8})$', model)
     if match:

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "autoforge-ai",
-  "version": "0.1.4",
+  "version": "0.1.5",
   "description": "Autonomous coding agent with web UI - build complete apps with AI",
   "license": "AGPL-3.0",
   "bin": {

package/registry.py

@@ -46,10 +46,16 @@ def _migrate_registry_dir() -> None:
 # Available models with display names
 # To add a new model: add an entry here with {"id": "model-id", "name": "Display Name"}
 AVAILABLE_MODELS = [
-    {"id": "claude-opus-4-5-20251101", "name": "Claude Opus 4.5"},
-    {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet 4.5"},
+    {"id": "claude-opus-4-6", "name": "Claude Opus"},
+    {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet"},
 ]
 
+# Map legacy model IDs to their current replacements.
+# Used by get_all_settings() to auto-migrate stale values on first read after upgrade.
+LEGACY_MODEL_MAP = {
+    "claude-opus-4-5-20251101": "claude-opus-4-6",
+}
+
 # List of valid model IDs (derived from AVAILABLE_MODELS)
 VALID_MODELS = [m["id"] for m in AVAILABLE_MODELS]
 
@@ -59,7 +65,7 @@ VALID_MODELS = [m["id"] for m in AVAILABLE_MODELS]
 _env_default_model = os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL")
 if _env_default_model is not None:
     _env_default_model = _env_default_model.strip()
-DEFAULT_MODEL = _env_default_model or "claude-opus-4-5-20251101"
+DEFAULT_MODEL = _env_default_model or "claude-opus-4-6"
 
 # Ensure env-provided DEFAULT_MODEL is in VALID_MODELS for validation consistency
 # (idempotent: only adds if missing, doesn't alter AVAILABLE_MODELS semantics)
@@ -598,6 +604,9 @@ def get_all_settings() -> dict[str, str]:
     """
     Get all settings as a dictionary.
 
+    Automatically migrates legacy model IDs (e.g. claude-opus-4-5-20251101 -> claude-opus-4-6)
+    on first read after upgrade. This is a one-time silent migration.
+
     Returns:
         Dictionary mapping setting keys to values.
     """
@@ -606,7 +615,26 @@ def get_all_settings() -> dict[str, str]:
         session = SessionLocal()
         try:
             settings = session.query(Settings).all()
-            return {s.key: s.value for s in settings}
+            result = {s.key: s.value for s in settings}
+
+            # Auto-migrate legacy model IDs
+            migrated = False
+            for key in ("model", "api_model"):
+                old_id = result.get(key)
+                if old_id and old_id in LEGACY_MODEL_MAP:
+                    new_id = LEGACY_MODEL_MAP[old_id]
+                    setting = session.query(Settings).filter(Settings.key == key).first()
+                    if setting:
+                        setting.value = new_id
+                        setting.updated_at = datetime.now()
+                        result[key] = new_id
+                        migrated = True
+                        logger.info("Migrated setting '%s': %s -> %s", key, old_id, new_id)
+
+            if migrated:
+                session.commit()
+
+            return result
         finally:
             session.close()
     except Exception as e:
@@ -624,10 +652,10 @@ API_PROVIDERS: dict[str, dict[str, Any]] = {
         "base_url": None,
         "requires_auth": False,
         "models": [
-            {"id": "claude-opus-4-5-20251101", "name": "Claude Opus 4.5"},
-            {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet 4.5"},
+            {"id": "claude-opus-4-6", "name": "Claude Opus"},
+            {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet"},
         ],
-        "default_model": "claude-opus-4-5-20251101",
+        "default_model": "claude-opus-4-6",
     },
     "kimi": {
         "name": "Kimi K2.5 (Moonshot)",

package/server/routers/settings.py

@@ -7,7 +7,6 @@ Settings are stored in the registry database and shared across all projects.
 """
 
 import mimetypes
-import os
 import sys
 
 from fastapi import APIRouter
@@ -39,19 +38,6 @@ def _parse_yolo_mode(value: str | None) -> bool:
     return (value or "false").lower() == "true"
 
 
-def _is_glm_mode() -> bool:
-    """Check if GLM API is configured via environment variables."""
-    base_url = os.getenv("ANTHROPIC_BASE_URL", "")
-    # GLM mode is when ANTHROPIC_BASE_URL is set but NOT pointing to Ollama
-    return bool(base_url) and not _is_ollama_mode()
-
-
-def _is_ollama_mode() -> bool:
-    """Check if Ollama API is configured via environment variables."""
-    base_url = os.getenv("ANTHROPIC_BASE_URL", "")
-    return "localhost:11434" in base_url or "127.0.0.1:11434" in base_url
-
-
 @router.get("/providers", response_model=ProvidersResponse)
 async def get_available_providers():
     """Get list of available API providers."""
@@ -116,9 +102,8 @@ async def get_settings():
 
     api_provider = all_settings.get("api_provider", "claude")
 
-    # Compute glm_mode / ollama_mode from api_provider for backward compat
-    glm_mode = api_provider == "glm" or _is_glm_mode()
-    ollama_mode = api_provider == "ollama" or _is_ollama_mode()
+    glm_mode = api_provider == "glm"
+    ollama_mode = api_provider == "ollama"
 
     return SettingsResponse(
         yolo_mode=_parse_yolo_mode(all_settings.get("yolo_mode")),
@@ -181,8 +166,8 @@ async def update_settings(update: SettingsUpdate):
     # Return updated settings
     all_settings = get_all_settings()
     api_provider = all_settings.get("api_provider", "claude")
-    glm_mode = api_provider == "glm" or _is_glm_mode()
-    ollama_mode = api_provider == "ollama" or _is_ollama_mode()
+    glm_mode = api_provider == "glm"
+    ollama_mode = api_provider == "ollama"
 
     return SettingsResponse(
         yolo_mode=_parse_yolo_mode(all_settings.get("yolo_mode")),

package/server/schemas.py

@@ -411,8 +411,8 @@ class SettingsResponse(BaseModel):
     """Response schema for global settings."""
     yolo_mode: bool = False
     model: str = DEFAULT_MODEL
-    glm_mode: bool = False  # True if GLM API is configured via .env
-    ollama_mode: bool = False  # True if Ollama API is configured via .env
+    glm_mode: bool = False  # True when api_provider is "glm"
+    ollama_mode: bool = False  # True when api_provider is "ollama"
     testing_agent_ratio: int = 1  # Regression testing agents (0-3)
     playwright_headless: bool = True
     batch_size: int = 3  # Features per coding agent batch (1-3)

package/server/services/assistant_chat_session.py

@@ -157,7 +157,7 @@ class AssistantChatSession:
     """
     Manages a read-only assistant conversation for a project.
 
-    Uses Claude Opus 4.5 with only read-only tools enabled.
+    Uses Claude Opus with only read-only tools enabled.
     Persists conversation history to SQLite.
     """
 
@@ -258,11 +258,11 @@ class AssistantChatSession:
         system_cli = shutil.which("claude")
 
         # Build environment overrides for API configuration
-        from registry import get_effective_sdk_env
+        from registry import DEFAULT_MODEL, get_effective_sdk_env
         sdk_env = get_effective_sdk_env()
 
         # Determine model from SDK env (provider-aware) or fallback to env/default
-        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", "claude-opus-4-5-20251101")
+        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
 
         try:
             logger.info("Creating ClaudeSDKClient...")

package/server/services/expand_chat_session.py

@@ -154,11 +154,11 @@ class ExpandChatSession:
         system_prompt = skill_content.replace("$ARGUMENTS", project_path)
 
         # Build environment overrides for API configuration
-        from registry import get_effective_sdk_env
+        from registry import DEFAULT_MODEL, get_effective_sdk_env
         sdk_env = get_effective_sdk_env()
 
         # Determine model from SDK env (provider-aware) or fallback to env/default
-        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", "claude-opus-4-5-20251101")
+        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
 
         # Build MCP servers config for feature creation
         mcp_servers = {

package/server/services/process_manager.py

@@ -346,7 +346,7 @@ class AgentProcessManager:
 
         Args:
             yolo_mode: If True, run in YOLO mode (skip testing agents)
-            model: Model to use (e.g., claude-opus-4-5-20251101)
+            model: Model to use (e.g., claude-opus-4-6)
             parallel_mode: DEPRECATED - ignored, always uses unified orchestrator
             max_concurrency: Max concurrent coding agents (1-5, default 1)
             testing_agent_ratio: Number of regression testing agents (0-3, default 1)

package/server/services/spec_chat_session.py

@@ -140,11 +140,11 @@ class SpecChatSession:
         system_cli = shutil.which("claude")
 
         # Build environment overrides for API configuration
-        from registry import get_effective_sdk_env
+        from registry import DEFAULT_MODEL, get_effective_sdk_env
         sdk_env = get_effective_sdk_env()
 
         # Determine model from SDK env (provider-aware) or fallback to env/default
-        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", "claude-opus-4-5-20251101")
+        model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
 
         try:
             self.client = ClaudeSDKClient(