layout.py: Support High-Level MCP

README.md

@@ -150,13 +150,15 @@ models:
   - krea/Krea-2-Turbo
   - lodestones/Chroma1-HD
   - lodestones/Chroma1-Radiance
+  - mage-flow-community/Mage-Flow
+  - mage-flow-community/Mage-Flow-Base
+  - mage-flow-community/Mage-Flow-Edit
+  - mage-flow-community/Mage-Flow-Edit-Base
+  - mage-flow-community/Mage-Flow-Edit-Turbo
+  - mage-flow-community/Mage-Flow-Turbo
   - meituan-longcat/LongCat-Image
   - microsoft/Lens
   - microsoft/Lens-Turbo
-  - microsoft/Mage-Flow
-  - microsoft/Mage-Flow-Edit
-  - microsoft/Mage-Flow-Edit-Turbo
-  - microsoft/Mage-Flow-Turbo
   - NewBie-AI/NewBie-image-Exp0.1
   - nvidia/PiD
   - nvidia/PixelDiT-1300M-1024px

mcp_tools/__init__.py

@@ -1,47 +1,47 @@
-def __getattr__(name):
-    if name in ("types", "server", "client", "shared"):
-        raise ImportError(f"No module named 'mcp.{name}' in local mcp package")
-    raise AttributeError(f"module '{__name__}' has no attribute '{name}'")
-
-from .get_task_list import handle_get_task_list
-from .get_model_architecture_list import handle_get_model_architecture_list
-from .get_model_list import handle_get_model_list
-from .get_feature_list import handle_get_feature_list
-from .get_model_features import handle_get_model_features
-from .run import handle_run
-from .get_task_status import handle_get_task_status
-from .error_schema import make_error, make_validation_error, make_not_found_error
-from .mcp_gradio_integration import (
-    register_high_level_mcp_apis,
-    cleanup_dependencies_api_names,
-    patch_gradio_api_suppression,
-    HIGH_LEVEL_MCP_API_NAMES,
-)
-
-MCP_FUNCTIONS = [
-    handle_get_task_list,
-    handle_get_model_architecture_list,
-    handle_get_model_list,
-    handle_get_feature_list,
-    handle_get_model_features,
-    handle_run,
-    handle_get_task_status,
-]
-
-__all__ = [
-    "handle_get_task_list",
-    "handle_get_model_architecture_list",
-    "handle_get_model_list",
-    "handle_get_feature_list",
-    "handle_get_model_features",
-    "handle_run",
-    "handle_get_task_status",
-    "make_error",
-    "make_validation_error",
-    "make_not_found_error",
-    "register_high_level_mcp_apis",
-    "cleanup_dependencies_api_names",
-    "patch_gradio_api_suppression",
-    "HIGH_LEVEL_MCP_API_NAMES",
-    "MCP_FUNCTIONS",
-]
+def __getattr__(name):
+    if name in ("types", "server", "client", "shared"):
+        raise ImportError(f"No module named 'mcp.{name}' in local mcp package")
+    raise AttributeError(f"module '{__name__}' has no attribute '{name}'")
+
+from .get_task_list import handle_get_task_list
+from .get_model_architecture_list import handle_get_model_architecture_list
+from .get_model_list import handle_get_model_list
+from .get_feature_list import handle_get_feature_list
+from .get_model_features import handle_get_model_features
+from .run import handle_run
+from .get_task_status import handle_get_task_status
+from .error_schema import make_error, make_validation_error, make_not_found_error
+from .mcp_gradio_integration import (
+    register_high_level_mcp_apis,
+    cleanup_dependencies_api_names,
+    patch_gradio_api_suppression,
+    HIGH_LEVEL_MCP_API_NAMES,
+)
+
+MCP_FUNCTIONS = [
+    handle_get_task_list,
+    handle_get_model_architecture_list,
+    handle_get_model_list,
+    handle_get_feature_list,
+    handle_get_model_features,
+    handle_run,
+    handle_get_task_status,
+]
+
+__all__ = [
+    "handle_get_task_list",
+    "handle_get_model_architecture_list",
+    "handle_get_model_list",
+    "handle_get_feature_list",
+    "handle_get_model_features",
+    "handle_run",
+    "handle_get_task_status",
+    "make_error",
+    "make_validation_error",
+    "make_not_found_error",
+    "register_high_level_mcp_apis",
+    "cleanup_dependencies_api_names",
+    "patch_gradio_api_suppression",
+    "HIGH_LEVEL_MCP_API_NAMES",
+    "MCP_FUNCTIONS",
+]

mcp_tools/error_schema.py

@@ -1,85 +1,85 @@
-"""
-Unified MCP tool error response format.
-
-Error code enumeration:
-  - INVALID_PARAMS:         Parameter validation failed (missing required fields, type errors, value out of range)
-  - MODEL_NOT_FOUND:        The specified model name does not exist
-  - ARCHITECTURE_NOT_FOUND: The specified architecture name does not exist
-  - CHAIN_TYPE_NOT_FOUND:   The specified chain/injector type is invalid
-  - FEATURE_NOT_SUPPORTED:  The current model does not support the requested feature
-  - TASK_NOT_FOUND:         The async task ID does not exist
-  - MODEL_OOM:              GPU out of memory
-  - INTERNAL_ERROR:         Internal server error
-"""
-
-
-def make_error(code: str, message: str, details: dict = None) -> dict:
-    """
-    Construct a unified MCP tool error response.
-
-    Args:
-        code:    Error code (UPPER_SNAKE_CASE format)
-        message: Human-readable error description
-        details: Optional details dictionary
-
-    Returns:
-        Standardized error response dictionary
-    """
-    error = {
-        "error": {
-            "code": code,
-            "message": message,
-        }
-    }
-    if details:
-        error["error"]["details"] = details
-    return error
-
-
-def make_validation_error(
-    message: str = "Request validation failed.",
-    missing_fields: list = None,
-    invalid_fields: dict = None,
-) -> dict:
-    """
-    Construct a parameter validation failure error response.
-
-    Args:
-        message:        Error description
-        missing_fields: List of missing required field names
-        invalid_fields: Key-value pairs of invalid fields, key=field name, value=reason description
-
-    Returns:
-        Standardized INVALID_PARAMS error response
-    """
-    details = {}
-    if missing_fields:
-        details["missing_fields"] = missing_fields
-    if invalid_fields:
-        details["invalid_fields"] = invalid_fields
-    return make_error("INVALID_PARAMS", message, details if details else None)
-
-
-def make_not_found_error(resource_type: str, resource_id: str) -> dict:
-    """
-    Construct a resource-not-found error response.
-
-    Args:
-        resource_type: Resource type (e.g., "model", "architecture", "chain_type", "task")
-        resource_id:   Resource identifier
-
-    Returns:
-        Standardized *_NOT_FOUND error response
-    """
-    code_map = {
-        "model": "MODEL_NOT_FOUND",
-        "architecture": "ARCHITECTURE_NOT_FOUND",
-        "chain_type": "CHAIN_TYPE_NOT_FOUND",
-        "task": "TASK_NOT_FOUND",
-    }
-    code = code_map.get(resource_type, f"{resource_type.upper()}_NOT_FOUND")
-    return make_error(
-        code,
-        f"The specified {resource_type} '{resource_id}' was not found.",
-        {"resource_type": resource_type, "resource_id": resource_id},
-    )
+"""
+Unified MCP tool error response format.
+
+Error code enumeration:
+  - INVALID_PARAMS:         Parameter validation failed (missing required fields, type errors, value out of range)
+  - MODEL_NOT_FOUND:        The specified model name does not exist
+  - ARCHITECTURE_NOT_FOUND: The specified architecture name does not exist
+  - CHAIN_TYPE_NOT_FOUND:   The specified chain/injector type is invalid
+  - FEATURE_NOT_SUPPORTED:  The current model does not support the requested feature
+  - TASK_NOT_FOUND:         The async task ID does not exist
+  - MODEL_OOM:              GPU out of memory
+  - INTERNAL_ERROR:         Internal server error
+"""
+
+
+def make_error(code: str, message: str, details: dict = None) -> dict:
+    """
+    Construct a unified MCP tool error response.
+
+    Args:
+        code:    Error code (UPPER_SNAKE_CASE format)
+        message: Human-readable error description
+        details: Optional details dictionary
+
+    Returns:
+        Standardized error response dictionary
+    """
+    error = {
+        "error": {
+            "code": code,
+            "message": message,
+        }
+    }
+    if details:
+        error["error"]["details"] = details
+    return error
+
+
+def make_validation_error(
+    message: str = "Request validation failed.",
+    missing_fields: list = None,
+    invalid_fields: dict = None,
+) -> dict:
+    """
+    Construct a parameter validation failure error response.
+
+    Args:
+        message:        Error description
+        missing_fields: List of missing required field names
+        invalid_fields: Key-value pairs of invalid fields, key=field name, value=reason description
+
+    Returns:
+        Standardized INVALID_PARAMS error response
+    """
+    details = {}
+    if missing_fields:
+        details["missing_fields"] = missing_fields
+    if invalid_fields:
+        details["invalid_fields"] = invalid_fields
+    return make_error("INVALID_PARAMS", message, details if details else None)
+
+
+def make_not_found_error(resource_type: str, resource_id: str) -> dict:
+    """
+    Construct a resource-not-found error response.
+
+    Args:
+        resource_type: Resource type (e.g., "model", "architecture", "chain_type", "task")
+        resource_id:   Resource identifier
+
+    Returns:
+        Standardized *_NOT_FOUND error response
+    """
+    code_map = {
+        "model": "MODEL_NOT_FOUND",
+        "architecture": "ARCHITECTURE_NOT_FOUND",
+        "chain_type": "CHAIN_TYPE_NOT_FOUND",
+        "task": "TASK_NOT_FOUND",
+    }
+    code = code_map.get(resource_type, f"{resource_type.upper()}_NOT_FOUND")
+    return make_error(
+        code,
+        f"The specified {resource_type} '{resource_id}' was not found.",
+        {"resource_type": resource_type, "resource_id": resource_id},
+    )

mcp_tools/get_feature_list.py

@@ -1,127 +1,127 @@
-import os
-from copy import deepcopy
-from .common import _load_yaml, _CHAIN_FEATURES_PATH, _YAML_DIR
-from .error_schema import make_not_found_error
-
-
-def _build_feature_entry(chain_name: str, chain_data: dict, include_schema: bool = False) -> dict:
-    entry = {
-        "feature_name": chain_name,
-        "chains": chain_data.get("chains", chain_name),
-        "display_name": chain_data.get("display_name", chain_name),
-        "description": chain_data.get("description", ""),
-        "supported_tasks": chain_data.get("supported_tasks", []),
-        "max_count": chain_data.get("max_count", 1),
-        "usage_guideline": chain_data.get("usage_guideline", ""),
-    }
-    if include_schema:
-        schema = deepcopy(chain_data.get("parameters_schema", {}))
-        if chain_name in ("krea2_controlnet", "diffsynth_controlnet", "controlnet", "anima_controlnet_lllite"):
-            config_key = (
-                "Krea2_ControlNet" if chain_name == "krea2_controlnet"
-                else "DiffSynth_ControlNet" if chain_name == "diffsynth_controlnet"
-                else "Anima_ControlNet_Lllite" if chain_name == "anima_controlnet_lllite"
-                else "ControlNet"
-            )
-            yaml_filename = f"{chain_name}_models.yaml"
-            model_path = os.path.join(_YAML_DIR, yaml_filename)
-            raw_models = _load_yaml(model_path).get(config_key, [])
-            models_list = []
-            if isinstance(raw_models, dict):
-                for val in raw_models.values():
-                    if isinstance(val, list):
-                        models_list.extend(val)
-                    elif isinstance(val, dict):
-                        models_list.append(val)
-            elif isinstance(raw_models, list):
-                models_list = raw_models
-
-            types_set = set()
-            for m in models_list:
-                t_val = m.get("Type", [])
-                if isinstance(t_val, list):
-                    types_set.update(t_val)
-                elif isinstance(t_val, str):
-                    types_set.add(t_val)
-            types = sorted(list(types_set))
-            series = sorted(list(set(m.get("Series") for m in models_list if m.get("Series"))))
-            if "properties" in schema:
-                if "type" in schema["properties"] and types:
-                    schema["properties"]["type"]["enum"] = types
-                if "series" in schema["properties"] and series:
-                    schema["properties"]["series"]["enum"] = series
-            if chain_name == "controlnet" and isinstance(raw_models, dict):
-                schema["architectures"] = raw_models
-        elif chain_name == "ipadapter":
-            from .common import _get_ipadapter_presets_by_arch
-            presets_by_arch = _get_ipadapter_presets_by_arch()
-            schema["presets_by_architecture"] = presets_by_arch
-            all_presets = sorted(list(set(presets_by_arch.get("SD1.5", []) + presets_by_arch.get("SDXL", []))))
-            if "properties" in schema and "preset" in schema["properties"]:
-                schema["properties"]["preset"]["enum"] = all_presets
-        entry["parameters_schema"] = schema
-    return entry
-
-
-def handle_get_feature_list(feature_name: str | list[str] = "") -> list | dict:
-    """
-    Dynamically load supported advanced features from chain_features.yaml.
-
-    - If feature_name is empty: returns a summary list of ALL features (excluding parameters_schema)
-      to optimize response size and token usage.
-    - If feature_name is specified (single feature name, comma-separated string, or list of strings):
-      returns complete feature details INCLUDING parameters_schema for the requested feature(s).
-    """
-    chain_features = _load_yaml(_CHAIN_FEATURES_PATH)
-
-    targets = []
-    is_single_string_query = False
-
-    if isinstance(feature_name, list):
-        targets = [str(x).strip() for x in feature_name if str(x).strip()]
-    elif isinstance(feature_name, str) and feature_name.strip():
-        raw_str = feature_name.strip()
-        parts = [x.strip() for x in raw_str.split(",") if x.strip()]
-        targets = parts
-        if len(parts) == 1 and "," not in raw_str:
-            is_single_string_query = True
-
-    # Case 1: Empty input -> return summary list of all features (without parameters_schema)
-    if not targets:
-        return [
-            _build_feature_entry(name, data, include_schema=False)
-            for name, data in chain_features.items()
-        ]
-
-    # Helper function to resolve feature target by key or chains alias
-    def _resolve_target(target_name: str) -> str | None:
-        if target_name in chain_features:
-            return target_name
-        for feat_key, feat_data in chain_features.items():
-            feat_chains = feat_data.get("chains")
-            if isinstance(feat_chains, str) and feat_chains == target_name:
-                return feat_key
-            elif isinstance(feat_chains, list) and target_name in feat_chains:
-                return feat_key
-        return None
-
-    resolved_targets = []
-    # Case 2: Specific feature(s) requested -> validate existence
-    for target in targets:
-        resolved = _resolve_target(target)
-        if not resolved:
-            return make_not_found_error("feature_name", target)
-        resolved_targets.append(resolved)
-
-    # Case 3: Return full info including parameters_schema
-    results = [
-        _build_feature_entry(target, chain_features[target], include_schema=True)
-        for target in resolved_targets
-    ]
-
-    if is_single_string_query and len(results) == 1:
-        return results[0]
-
-    return results
-
-
+import os
+from copy import deepcopy
+from .common import _load_yaml, _CHAIN_FEATURES_PATH, _YAML_DIR
+from .error_schema import make_not_found_error
+
+
+def _build_feature_entry(chain_name: str, chain_data: dict, include_schema: bool = False) -> dict:
+    entry = {
+        "feature_name": chain_name,
+        "chains": chain_data.get("chains", chain_name),
+        "display_name": chain_data.get("display_name", chain_name),
+        "description": chain_data.get("description", ""),
+        "supported_tasks": chain_data.get("supported_tasks", []),
+        "max_count": chain_data.get("max_count", 1),
+        "usage_guideline": chain_data.get("usage_guideline", ""),
+    }
+    if include_schema:
+        schema = deepcopy(chain_data.get("parameters_schema", {}))
+        if chain_name in ("krea2_controlnet", "diffsynth_controlnet", "controlnet", "anima_controlnet_lllite"):
+            config_key = (
+                "Krea2_ControlNet" if chain_name == "krea2_controlnet"
+                else "DiffSynth_ControlNet" if chain_name == "diffsynth_controlnet"
+                else "Anima_ControlNet_Lllite" if chain_name == "anima_controlnet_lllite"
+                else "ControlNet"
+            )
+            yaml_filename = f"{chain_name}_models.yaml"
+            model_path = os.path.join(_YAML_DIR, yaml_filename)
+            raw_models = _load_yaml(model_path).get(config_key, [])
+            models_list = []
+            if isinstance(raw_models, dict):
+                for val in raw_models.values():
+                    if isinstance(val, list):
+                        models_list.extend(val)
+                    elif isinstance(val, dict):
+                        models_list.append(val)
+            elif isinstance(raw_models, list):
+                models_list = raw_models
+
+            types_set = set()
+            for m in models_list:
+                t_val = m.get("Type", [])
+                if isinstance(t_val, list):
+                    types_set.update(t_val)
+                elif isinstance(t_val, str):
+                    types_set.add(t_val)
+            types = sorted(list(types_set))
+            series = sorted(list(set(m.get("Series") for m in models_list if m.get("Series"))))
+            if "properties" in schema:
+                if "type" in schema["properties"] and types:
+                    schema["properties"]["type"]["enum"] = types
+                if "series" in schema["properties"] and series:
+                    schema["properties"]["series"]["enum"] = series
+            if chain_name == "controlnet" and isinstance(raw_models, dict):
+                schema["architectures"] = raw_models
+        elif chain_name == "ipadapter":
+            from .common import _get_ipadapter_presets_by_arch
+            presets_by_arch = _get_ipadapter_presets_by_arch()
+            schema["presets_by_architecture"] = presets_by_arch
+            all_presets = sorted(list(set(presets_by_arch.get("SD1.5", []) + presets_by_arch.get("SDXL", []))))
+            if "properties" in schema and "preset" in schema["properties"]:
+                schema["properties"]["preset"]["enum"] = all_presets
+        entry["parameters_schema"] = schema
+    return entry
+
+
+def handle_get_feature_list(feature_name: str | list[str] = "") -> list | dict:
+    """
+    Dynamically load supported advanced features from chain_features.yaml.
+
+    - If feature_name is empty: returns a summary list of ALL features (excluding parameters_schema)
+      to optimize response size and token usage.
+    - If feature_name is specified (single feature name, comma-separated string, or list of strings):
+      returns complete feature details INCLUDING parameters_schema for the requested feature(s).
+    """
+    chain_features = _load_yaml(_CHAIN_FEATURES_PATH)
+
+    targets = []
+    is_single_string_query = False
+
+    if isinstance(feature_name, list):
+        targets = [str(x).strip() for x in feature_name if str(x).strip()]
+    elif isinstance(feature_name, str) and feature_name.strip():
+        raw_str = feature_name.strip()
+        parts = [x.strip() for x in raw_str.split(",") if x.strip()]
+        targets = parts
+        if len(parts) == 1 and "," not in raw_str:
+            is_single_string_query = True
+
+    # Case 1: Empty input -> return summary list of all features (without parameters_schema)
+    if not targets:
+        return [
+            _build_feature_entry(name, data, include_schema=False)
+            for name, data in chain_features.items()
+        ]
+
+    # Helper function to resolve feature target by key or chains alias
+    def _resolve_target(target_name: str) -> str | None:
+        if target_name in chain_features:
+            return target_name
+        for feat_key, feat_data in chain_features.items():
+            feat_chains = feat_data.get("chains")
+            if isinstance(feat_chains, str) and feat_chains == target_name:
+                return feat_key
+            elif isinstance(feat_chains, list) and target_name in feat_chains:
+                return feat_key
+        return None
+
+    resolved_targets = []
+    # Case 2: Specific feature(s) requested -> validate existence
+    for target in targets:
+        resolved = _resolve_target(target)
+        if not resolved:
+            return make_not_found_error("feature_name", target)
+        resolved_targets.append(resolved)
+
+    # Case 3: Return full info including parameters_schema
+    results = [
+        _build_feature_entry(target, chain_features[target], include_schema=True)
+        for target in resolved_targets
+    ]
+
+    if is_single_string_query and len(results) == 1:
+        return results[0]
+
+    return results
+
+

mcp_tools/get_model_architecture_list.py

@@ -1,36 +1,36 @@
-"""
-MCP Tool: get_model_architecture_list
-Get all supported model architectures and their corresponding default resolutions.
-"""
-
-from .common import _load_yaml, _MODEL_ARCHITECTURES_PATH, _CONSTANTS_PATH
-
-
-def handle_get_model_architecture_list() -> list:
-    """Dynamically load all supported model architectures from model_architectures.yaml."""
-    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
-    constants = _load_yaml(_CONSTANTS_PATH)
-    resolution_map = constants.get("RESOLUTION_MAP", {})
-    architectures = arch_config.get("architectures", {})
-    architecture_order = arch_config.get("architecture_order", list(architectures.keys()))
-
-    result = []
-    for arch_name in architecture_order:
-        if arch_name not in architectures:
-            continue
-        arch_data = architectures[arch_name]
-        model_type = arch_data.get("model_type", arch_name.lower())
-
-        default_res = [1024, 1024]
-        if model_type in resolution_map:
-            resolutions = resolution_map[model_type]
-            if resolutions:
-                first_key = next(iter(resolutions))
-                default_res = resolutions[first_key]
-
-        result.append({
-            "model_architecture": arch_name,
-            "default_resolution": default_res,
-        })
-
-    return result
+"""
+MCP Tool: get_model_architecture_list
+Get all supported model architectures and their corresponding default resolutions.
+"""
+
+from .common import _load_yaml, _MODEL_ARCHITECTURES_PATH, _CONSTANTS_PATH
+
+
+def handle_get_model_architecture_list() -> list:
+    """Dynamically load all supported model architectures from model_architectures.yaml."""
+    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
+    constants = _load_yaml(_CONSTANTS_PATH)
+    resolution_map = constants.get("RESOLUTION_MAP", {})
+    architectures = arch_config.get("architectures", {})
+    architecture_order = arch_config.get("architecture_order", list(architectures.keys()))
+
+    result = []
+    for arch_name in architecture_order:
+        if arch_name not in architectures:
+            continue
+        arch_data = architectures[arch_name]
+        model_type = arch_data.get("model_type", arch_name.lower())
+
+        default_res = [1024, 1024]
+        if model_type in resolution_map:
+            resolutions = resolution_map[model_type]
+            if resolutions:
+                first_key = next(iter(resolutions))
+                default_res = resolutions[first_key]
+
+        result.append({
+            "model_architecture": arch_name,
+            "default_resolution": default_res,
+        })
+
+    return result

mcp_tools/get_model_features.py

@@ -1,102 +1,102 @@
-"""
-MCP Tool: get_model_features
-Query metadata for a specified model, including supported task types, extended features, and default inference parameters.
-"""
-
-from .common import (
-    _load_yaml,
-    _MODEL_LIST_PATH,
-    _MODEL_DEFAULTS_PATH,
-    _IMAGE_GEN_FEATURES_PATH,
-    _MODEL_ARCHITECTURES_PATH,
-    _CHAIN_FEATURES_PATH,
-    _TASK_DEFINITIONS,
-)
-from .error_schema import make_validation_error, make_not_found_error
-
-
-def handle_get_model_features(model: str) -> dict:
-    """Query metadata for a specified model: supported task types, extended features, and default inference parameters."""
-    if not model:
-        return make_validation_error(
-            "Parameter 'model' is required.",
-            missing_fields=["model"],
-        )
-
-    model_list = _load_yaml(_MODEL_LIST_PATH)
-    model_defaults = _load_yaml(_MODEL_DEFAULTS_PATH)
-    features_config = _load_yaml(_IMAGE_GEN_FEATURES_PATH)
-    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
-    chain_features = _load_yaml(_CHAIN_FEATURES_PATH)
-
-    found_arch = None
-    checkpoints = model_list.get("Checkpoint", {})
-    for arch_name, arch_data in checkpoints.items():
-        if not isinstance(arch_data, dict):
-            continue
-        for m in arch_data.get("models", []):
-            if m.get("display_name") == model:
-                found_arch = arch_name
-                break
-        if found_arch:
-            break
-
-    if not found_arch:
-        return make_not_found_error("model", model)
-
-    architectures = arch_config.get("architectures", {})
-    arch_info = architectures.get(found_arch, {})
-    model_type = arch_info.get("model_type", found_arch.lower())
-
-    arch_features = features_config.get(model_type, features_config.get("default", {}))
-    enabled_chains = arch_features.get("enabled_chains", [])
-
-    supported_features = []
-    for feat_name, feat_data in chain_features.items():
-        feat_chains = feat_data.get("chains")
-        if feat_chains is None:
-            feat_chains = [feat_name]
-        elif isinstance(feat_chains, str):
-            feat_chains = [feat_chains]
-
-        if any(c in enabled_chains for c in feat_chains):
-            supported_features.append(feat_name)
-
-    arch_defaults_section = model_defaults.get(found_arch, {})
-    arch_level_defaults = arch_defaults_section.get("_defaults", {})
-    model_specific_defaults = arch_defaults_section.get(model, {})
-    global_defaults = model_defaults.get("Default", {})
-
-    merged_defaults = {**global_defaults, **arch_level_defaults, **model_specific_defaults}
-
-    default_parameter = {
-        "sampler": merged_defaults.get("sampler_name", "euler"),
-        "scheduler": merged_defaults.get("scheduler", "simple"),
-        "steps": merged_defaults.get("steps", 20),
-        "cfg": merged_defaults.get("cfg", 1.0),
-    }
-
-    supported_tasks = [t["task_type"] for t in _TASK_DEFINITIONS]
-
-    result = {
-        "name": model,
-        "model_architecture": found_arch,
-        "supported_tasks": supported_tasks,
-        "supported_features": supported_features,
-        "default_parameter": default_parameter,
-    }
-
-    default_pos = model_specific_defaults.get(
-        "positive_prompt",
-        arch_level_defaults.get("positive_prompt", ""),
-    )
-    default_neg = model_specific_defaults.get(
-        "negative_prompt",
-        arch_level_defaults.get("negative_prompt", ""),
-    )
-    if default_pos:
-        result["default_positive_prompt"] = default_pos
-    if default_neg:
-        result["default_negative_prompt"] = default_neg
-
-    return result
+"""
+MCP Tool: get_model_features
+Query metadata for a specified model, including supported task types, extended features, and default inference parameters.
+"""
+
+from .common import (
+    _load_yaml,
+    _MODEL_LIST_PATH,
+    _MODEL_DEFAULTS_PATH,
+    _IMAGE_GEN_FEATURES_PATH,
+    _MODEL_ARCHITECTURES_PATH,
+    _CHAIN_FEATURES_PATH,
+    _TASK_DEFINITIONS,
+)
+from .error_schema import make_validation_error, make_not_found_error
+
+
+def handle_get_model_features(model: str) -> dict:
+    """Query metadata for a specified model: supported task types, extended features, and default inference parameters."""
+    if not model:
+        return make_validation_error(
+            "Parameter 'model' is required.",
+            missing_fields=["model"],
+        )
+
+    model_list = _load_yaml(_MODEL_LIST_PATH)
+    model_defaults = _load_yaml(_MODEL_DEFAULTS_PATH)
+    features_config = _load_yaml(_IMAGE_GEN_FEATURES_PATH)
+    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
+    chain_features = _load_yaml(_CHAIN_FEATURES_PATH)
+
+    found_arch = None
+    checkpoints = model_list.get("Checkpoint", {})
+    for arch_name, arch_data in checkpoints.items():
+        if not isinstance(arch_data, dict):
+            continue
+        for m in arch_data.get("models", []):
+            if m.get("display_name") == model:
+                found_arch = arch_name
+                break
+        if found_arch:
+            break
+
+    if not found_arch:
+        return make_not_found_error("model", model)
+
+    architectures = arch_config.get("architectures", {})
+    arch_info = architectures.get(found_arch, {})
+    model_type = arch_info.get("model_type", found_arch.lower())
+
+    arch_features = features_config.get(model_type, features_config.get("default", {}))
+    enabled_chains = arch_features.get("enabled_chains", [])
+
+    supported_features = []
+    for feat_name, feat_data in chain_features.items():
+        feat_chains = feat_data.get("chains")
+        if feat_chains is None:
+            feat_chains = [feat_name]
+        elif isinstance(feat_chains, str):
+            feat_chains = [feat_chains]
+
+        if any(c in enabled_chains for c in feat_chains):
+            supported_features.append(feat_name)
+
+    arch_defaults_section = model_defaults.get(found_arch, {})
+    arch_level_defaults = arch_defaults_section.get("_defaults", {})
+    model_specific_defaults = arch_defaults_section.get(model, {})
+    global_defaults = model_defaults.get("Default", {})
+
+    merged_defaults = {**global_defaults, **arch_level_defaults, **model_specific_defaults}
+
+    default_parameter = {
+        "sampler": merged_defaults.get("sampler_name", "euler"),
+        "scheduler": merged_defaults.get("scheduler", "simple"),
+        "steps": merged_defaults.get("steps", 20),
+        "cfg": merged_defaults.get("cfg", 1.0),
+    }
+
+    supported_tasks = [t["task_type"] for t in _TASK_DEFINITIONS]
+
+    result = {
+        "name": model,
+        "model_architecture": found_arch,
+        "supported_tasks": supported_tasks,
+        "supported_features": supported_features,
+        "default_parameter": default_parameter,
+    }
+
+    default_pos = model_specific_defaults.get(
+        "positive_prompt",
+        arch_level_defaults.get("positive_prompt", ""),
+    )
+    default_neg = model_specific_defaults.get(
+        "negative_prompt",
+        arch_level_defaults.get("negative_prompt", ""),
+    )
+    if default_pos:
+        result["default_positive_prompt"] = default_pos
+    if default_neg:
+        result["default_negative_prompt"] = default_neg
+
+    return result

mcp_tools/get_model_list.py

@@ -1,64 +1,64 @@
-"""
-MCP Tool: get_model_list
-Query the list of available image generation models, with optional filtering by model architecture.
-"""
-
-from .common import _load_yaml, _MODEL_LIST_PATH, _MODEL_DEFAULTS_PATH, _MODEL_ARCHITECTURES_PATH
-from .error_schema import make_not_found_error
-
-
-def handle_get_model_list(model_architecture: str = None) -> list | dict:
-    """Dynamically load the list of available image generation models from model_list.yaml."""
-    model_list = _load_yaml(_MODEL_LIST_PATH)
-    model_defaults = _load_yaml(_MODEL_DEFAULTS_PATH)
-    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
-    valid_architectures = set(arch_config.get("architectures", {}).keys())
-
-    if model_architecture and model_architecture not in valid_architectures:
-        return make_not_found_error("architecture", model_architecture)
-
-    result = []
-    checkpoints = model_list.get("Checkpoint", {})
-
-    for arch_name, arch_data in checkpoints.items():
-        if model_architecture and arch_name != model_architecture:
-            continue
-        if not isinstance(arch_data, dict):
-            continue
-
-        models = arch_data.get("models", [])
-        if not isinstance(models, list):
-            continue
-
-        arch_defaults = model_defaults.get(arch_name, {})
-        arch_level_defaults = arch_defaults.get("_defaults", {})
-
-        for model in models:
-            display_name = model.get("display_name", "")
-            category = model.get("category", None)
-
-            model_specific_defaults = arch_defaults.get(display_name, {})
-
-            default_pos = model_specific_defaults.get(
-                "positive_prompt",
-                arch_level_defaults.get("positive_prompt", ""),
-            )
-            default_neg = model_specific_defaults.get(
-                "negative_prompt",
-                arch_level_defaults.get("negative_prompt", ""),
-            )
-
-            entry = {
-                "name": display_name,
-                "model_architecture": arch_name,
-            }
-            if category:
-                entry["category"] = category
-            if default_pos:
-                entry["default_positive_prompt"] = default_pos
-            if default_neg:
-                entry["default_negative_prompt"] = default_neg
-
-            result.append(entry)
-
-    return result
+"""
+MCP Tool: get_model_list
+Query the list of available image generation models, with optional filtering by model architecture.
+"""
+
+from .common import _load_yaml, _MODEL_LIST_PATH, _MODEL_DEFAULTS_PATH, _MODEL_ARCHITECTURES_PATH
+from .error_schema import make_not_found_error
+
+
+def handle_get_model_list(model_architecture: str = None) -> list | dict:
+    """Dynamically load the list of available image generation models from model_list.yaml."""
+    model_list = _load_yaml(_MODEL_LIST_PATH)
+    model_defaults = _load_yaml(_MODEL_DEFAULTS_PATH)
+    arch_config = _load_yaml(_MODEL_ARCHITECTURES_PATH)
+    valid_architectures = set(arch_config.get("architectures", {}).keys())
+
+    if model_architecture and model_architecture not in valid_architectures:
+        return make_not_found_error("architecture", model_architecture)
+
+    result = []
+    checkpoints = model_list.get("Checkpoint", {})
+
+    for arch_name, arch_data in checkpoints.items():
+        if model_architecture and arch_name != model_architecture:
+            continue
+        if not isinstance(arch_data, dict):
+            continue
+
+        models = arch_data.get("models", [])
+        if not isinstance(models, list):
+            continue
+
+        arch_defaults = model_defaults.get(arch_name, {})
+        arch_level_defaults = arch_defaults.get("_defaults", {})
+
+        for model in models:
+            display_name = model.get("display_name", "")
+            category = model.get("category", None)
+
+            model_specific_defaults = arch_defaults.get(display_name, {})
+
+            default_pos = model_specific_defaults.get(
+                "positive_prompt",
+                arch_level_defaults.get("positive_prompt", ""),
+            )
+            default_neg = model_specific_defaults.get(
+                "negative_prompt",
+                arch_level_defaults.get("negative_prompt", ""),
+            )
+
+            entry = {
+                "name": display_name,
+                "model_architecture": arch_name,
+            }
+            if category:
+                entry["category"] = category
+            if default_pos:
+                entry["default_positive_prompt"] = default_pos
+            if default_neg:
+                entry["default_negative_prompt"] = default_neg
+
+            result.append(entry)
+
+    return result

mcp_tools/get_task_list.py

@@ -1,11 +1,11 @@
-"""
-MCP Tool: get_task_list
-Get a list of all supported image generation task types along with their required/optional parameter lists.
-"""
-
-from .common import _TASK_DEFINITIONS
-
-
-def handle_get_task_list() -> list:
-    """Get a list of all supported image generation task types along with their required/optional parameter lists."""
-    return _TASK_DEFINITIONS
+"""
+MCP Tool: get_task_list
+Get a list of all supported image generation task types along with their required/optional parameter lists.
+"""
+
+from .common import _TASK_DEFINITIONS
+
+
+def handle_get_task_list() -> list:
+    """Get a list of all supported image generation task types along with their required/optional parameter lists."""
+    return _TASK_DEFINITIONS

mcp_tools/get_task_status.py

@@ -1,21 +1,21 @@
-"""
-MCP Tool: get_task_status
-Query the processing progress and final results of an async image generation task.
-"""
-
-from .common import _TASKS_DB
-from .error_schema import make_validation_error, make_not_found_error
-
-
-def handle_get_task_status(task_id: str) -> dict:
-    """Query the processing progress and final results of an async image generation task."""
-    if not task_id:
-        return make_validation_error(
-            "Parameter 'task_id' is required.",
-            missing_fields=["task_id"],
-        )
-
-    if task_id not in _TASKS_DB:
-        return make_not_found_error("task", task_id)
-
-    return _TASKS_DB[task_id]
+"""
+MCP Tool: get_task_status
+Query the processing progress and final results of an async image generation task.
+"""
+
+from .common import _TASKS_DB
+from .error_schema import make_validation_error, make_not_found_error
+
+
+def handle_get_task_status(task_id: str) -> dict:
+    """Query the processing progress and final results of an async image generation task."""
+    if not task_id:
+        return make_validation_error(
+            "Parameter 'task_id' is required.",
+            missing_fields=["task_id"],
+        )
+
+    if task_id not in _TASKS_DB:
+        return make_not_found_error("task", task_id)
+
+    return _TASKS_DB[task_id]

mcp_tools/mcp_gradio_integration.py

@@ -1,130 +1,130 @@
-"""
-MCP & Gradio Integration Module
-
-Provides:
-1. register_high_level_mcp_apis: Expose only 7 high-level abstract API/MCP endpoints (using gr.api without polluting the visual UI structure)
-2. cleanup_dependencies_api_names: Force cleanup of show_api attribute for non-high-level APIs in dependencies
-3. patch_gradio_api_suppression: No-op implementation retained for backward compatibility
-"""
-
-import json
-import gradio as gr
-
-from .get_task_list import handle_get_task_list
-from .get_model_architecture_list import handle_get_model_architecture_list
-from .get_model_list import handle_get_model_list
-from .get_feature_list import handle_get_feature_list
-from .get_model_features import handle_get_model_features
-from .run import handle_run
-from .get_task_status import handle_get_task_status
-
-HIGH_LEVEL_MCP_API_NAMES = {
-    "get_task_list",
-    "get_model_architecture_list",
-    "get_model_list",
-    "get_feature_list",
-    "get_model_features",
-    "run",
-    "get_task_status",
-}
-
-
-def sanitize_keys(obj):
-    """Recursively ensure all dictionary keys are converted to str type to avoid Gradio 5 orjson TypeError: Dict key must be str."""
-    if isinstance(obj, dict):
-        return {str(k): sanitize_keys(v) for k, v in obj.items()}
-    elif isinstance(obj, list):
-        return [sanitize_keys(x) for x in obj]
-    elif isinstance(obj, tuple):
-        return tuple(sanitize_keys(x) for x in obj)
-    return obj
-
-
-def patch_gradio_api_suppression():
-    """Retained for backward compatibility (no-op)."""
-    pass
-
-
-def cleanup_dependencies_api_names(demo):
-    """
-    Clean up residual auto-generated API names in demo.fns and demo.dependencies.
-    Force only the 7 high-level abstract MCP APIs to be exposed as public endpoints.
-    """
-    for fn in demo.fns.values():
-        api_name = getattr(fn, "api_name", None)
-        if api_name not in HIGH_LEVEL_MCP_API_NAMES:
-            fn.show_api = False
-
-    deps = getattr(demo, "dependencies", None)
-    if deps is None and hasattr(demo, "config") and isinstance(demo.config, dict):
-        deps = demo.config.get("dependencies", [])
-
-    if deps:
-        for dep in deps:
-            if isinstance(dep, dict):
-                api_name = dep.get("api_name")
-                if api_name not in HIGH_LEVEL_MCP_API_NAMES:
-                    dep["show_api"] = False
-
-    print("[MCP Protection] Cleaned up demo dependencies. Suppressed atomic API endpoints.")
-
-
-def register_high_level_mcp_apis(demo):
-    """
-    Explicitly register 7 high-level abstract MCP API endpoints on the Gradio demo using gr.api.
-    Using gr.api() never adds any visual UI components (such as Row, Textbox, Button, etc.), avoiding duplicate interface rendering.
-    """
-    def get_task_list() -> list:
-        """[Recommended Discovery Flow Step 1] Get a list of all supported image generation task types (txt2img, img2img, inpaint, outpaint, hires_fix) along with their required and optional parameter lists. Recommended flow: get_task_list -> get_model_architecture_list -> get_model_list -> [Path 1: Call run directly (pass only required params) | Path 2: Call get_model_features to get official default hyperparams -> run]."""
-        return sanitize_keys(handle_get_task_list())
-
-    def get_model_architecture_list() -> list:
-        """[Recommended Discovery Flow Step 2] Get a list of all supported model architectures (e.g., SD1.5, SDXL, FLUX, etc.) along with their default resolutions. It is recommended to call this tool before get_model_list to obtain valid model_architecture parameters for precise model filtering."""
-        return sanitize_keys(handle_get_model_architecture_list())
-
-    def get_model_list(model_architecture: str = "") -> list | dict:
-        """[Recommended Discovery Flow Step 3] Query the list of available image generation models. After obtaining models, choose one of two paths: 1. [Path 1 (Recommended - Minimal Mode)] Call run directly with only required parameters. Do NOT guess steps/cfg/sampler/scheduler from experience; the server will automatically apply the model's optimal default hyperparameters. 2. [Path 2 (Explicit Alignment Mode)] First call get_model_features to query the model's officially recommended hyperparameters, then pass them to run."""
-        arch = model_architecture.strip() if model_architecture else None
-        return sanitize_keys(handle_get_model_list(arch))
-
-    def get_feature_list(feature_name: str = "") -> list | dict:
-        """Get supported advanced features. If feature_name is empty, returns a summary list of ALL features (excluding parameters_schema to save tokens). Pass a specific feature_name (single name like 'lora', or comma-separated like 'lora, ipadapter') to retrieve complete details INCLUDING parameters_schema for requested feature(s)."""
-        return sanitize_keys(handle_get_feature_list(feature_name.strip() if isinstance(feature_name, str) else feature_name))
-
-    def get_model_features(model: str = "") -> dict:
-        """Query metadata for the specified model, including supported task types, extended features, and official default inference parameters (steps, cfg, sampler, scheduler). This tool MUST be called when explicitly obtaining a model's optimal default hyperparameters (Path 2). Guessing or fabricating hyperparameters without querying is strictly prohibited."""
-        return sanitize_keys(handle_get_model_features(model.strip()))
-
-    def run(json_params: str = "{}") -> dict:
-        """[Recommended Discovery Flow Step 4] Unified image generation task execution interface. Supports txt2img, img2img, and other tasks with chainable extended features. [IMPORTANT PARAMETER RULES] Do NOT guess or fabricate inference hyperparameters such as steps, cfg, sampler, scheduler! Path 1 (Recommended): Pass only required parameters (task_type, model, prompt, width, height), leave optional hyperparams empty (server uses optimal defaults). Path 2: If explicit hyperparams are needed, you MUST first call get_model_features to obtain official defaults before passing them."""
-        try:
-            if isinstance(json_params, dict):
-                params = json_params
-            else:
-                params = json.loads(json_params or "{}")
-        except Exception as e:
-            return {"error": {"code": "INVALID_JSON", "message": f"Failed to parse JSON params: {e}"}}
-        return sanitize_keys(handle_run(params))
-
-    def get_task_status(task_id: str = "") -> dict:
-        """Query the progress, status, and final generated results of an async image generation task."""
-        return sanitize_keys(handle_get_task_status(task_id.strip()))
-
-    funcs = [
-        get_task_list,
-        get_model_architecture_list,
-        get_model_list,
-        get_feature_list,
-        get_model_features,
-        run,
-        get_task_status,
-    ]
-
-    for func in funcs:
-        gr.api(func)
-
-    for fn in demo.fns.values():
-        if getattr(fn, "api_name", None) in HIGH_LEVEL_MCP_API_NAMES:
-            fn.show_api = True
-
-    print("[MCP Integration] Successfully registered 7 High-Level Abstract MCP APIs via gr.api().")
+"""
+MCP & Gradio Integration Module
+
+Provides:
+1. register_high_level_mcp_apis: Expose only 7 high-level abstract API/MCP endpoints (using gr.api without polluting the visual UI structure)
+2. cleanup_dependencies_api_names: Force cleanup of show_api attribute for non-high-level APIs in dependencies
+3. patch_gradio_api_suppression: No-op implementation retained for backward compatibility
+"""
+
+import json
+import gradio as gr
+
+from .get_task_list import handle_get_task_list
+from .get_model_architecture_list import handle_get_model_architecture_list
+from .get_model_list import handle_get_model_list
+from .get_feature_list import handle_get_feature_list
+from .get_model_features import handle_get_model_features
+from .run import handle_run
+from .get_task_status import handle_get_task_status
+
+HIGH_LEVEL_MCP_API_NAMES = {
+    "get_task_list",
+    "get_model_architecture_list",
+    "get_model_list",
+    "get_feature_list",
+    "get_model_features",
+    "run",
+    "get_task_status",
+}
+
+
+def sanitize_keys(obj):
+    """Recursively ensure all dictionary keys are converted to str type to avoid Gradio 5 orjson TypeError: Dict key must be str."""
+    if isinstance(obj, dict):
+        return {str(k): sanitize_keys(v) for k, v in obj.items()}
+    elif isinstance(obj, list):
+        return [sanitize_keys(x) for x in obj]
+    elif isinstance(obj, tuple):
+        return tuple(sanitize_keys(x) for x in obj)
+    return obj
+
+
+def patch_gradio_api_suppression():
+    """Retained for backward compatibility (no-op)."""
+    pass
+
+
+def cleanup_dependencies_api_names(demo):
+    """
+    Clean up residual auto-generated API names in demo.fns and demo.dependencies.
+    Force only the 7 high-level abstract MCP APIs to be exposed as public endpoints.
+    """
+    for fn in demo.fns.values():
+        api_name = getattr(fn, "api_name", None)
+        if api_name not in HIGH_LEVEL_MCP_API_NAMES:
+            fn.show_api = False
+
+    deps = getattr(demo, "dependencies", None)
+    if deps is None and hasattr(demo, "config") and isinstance(demo.config, dict):
+        deps = demo.config.get("dependencies", [])
+
+    if deps:
+        for dep in deps:
+            if isinstance(dep, dict):
+                api_name = dep.get("api_name")
+                if api_name not in HIGH_LEVEL_MCP_API_NAMES:
+                    dep["show_api"] = False
+
+    print("[MCP Protection] Cleaned up demo dependencies. Suppressed atomic API endpoints.")
+
+
+def register_high_level_mcp_apis(demo):
+    """
+    Explicitly register 7 high-level abstract MCP API endpoints on the Gradio demo using gr.api.
+    Using gr.api() never adds any visual UI components (such as Row, Textbox, Button, etc.), avoiding duplicate interface rendering.
+    """
+    def get_task_list() -> list:
+        """[Recommended Discovery Flow Step 1] Get a list of all supported image generation task types (txt2img, img2img, inpaint, outpaint, hires_fix) along with their required and optional parameter lists. Recommended flow: get_task_list -> get_model_architecture_list -> get_model_list -> [Path 1: Call run directly (pass only required params) | Path 2: Call get_model_features to get official default hyperparams -> run]."""
+        return sanitize_keys(handle_get_task_list())
+
+    def get_model_architecture_list() -> list:
+        """[Recommended Discovery Flow Step 2] Get a list of all supported model architectures (e.g., SD1.5, SDXL, FLUX, etc.) along with their default resolutions. It is recommended to call this tool before get_model_list to obtain valid model_architecture parameters for precise model filtering."""
+        return sanitize_keys(handle_get_model_architecture_list())
+
+    def get_model_list(model_architecture: str = "") -> list | dict:
+        """[Recommended Discovery Flow Step 3] Query the list of available image generation models. After obtaining models, choose one of two paths: 1. [Path 1 (Recommended - Minimal Mode)] Call run directly with only required parameters. Do NOT guess steps/cfg/sampler/scheduler from experience; the server will automatically apply the model's optimal default hyperparameters. 2. [Path 2 (Explicit Alignment Mode)] First call get_model_features to query the model's officially recommended hyperparameters, then pass them to run."""
+        arch = model_architecture.strip() if model_architecture else None
+        return sanitize_keys(handle_get_model_list(arch))
+
+    def get_feature_list(feature_name: str = "") -> list | dict:
+        """Get supported advanced features. If feature_name is empty, returns a summary list of ALL features (excluding parameters_schema to save tokens). Pass a specific feature_name (single name like 'lora', or comma-separated like 'lora, ipadapter') to retrieve complete details INCLUDING parameters_schema for requested feature(s)."""
+        return sanitize_keys(handle_get_feature_list(feature_name.strip() if isinstance(feature_name, str) else feature_name))
+
+    def get_model_features(model: str = "") -> dict:
+        """Query metadata for the specified model, including supported task types, extended features, and official default inference parameters (steps, cfg, sampler, scheduler). This tool MUST be called when explicitly obtaining a model's optimal default hyperparameters (Path 2). Guessing or fabricating hyperparameters without querying is strictly prohibited."""
+        return sanitize_keys(handle_get_model_features(model.strip()))
+
+    def run(json_params: str = "{}") -> dict:
+        """[Recommended Discovery Flow Step 4] Unified image generation task execution interface. Supports txt2img, img2img, and other tasks with chainable extended features. [IMPORTANT PARAMETER RULES] Do NOT guess or fabricate inference hyperparameters such as steps, cfg, sampler, scheduler! Path 1 (Recommended): Pass only required parameters (task_type, model, prompt, width, height), leave optional hyperparams empty (server uses optimal defaults). Path 2: If explicit hyperparams are needed, you MUST first call get_model_features to obtain official defaults before passing them."""
+        try:
+            if isinstance(json_params, dict):
+                params = json_params
+            else:
+                params = json.loads(json_params or "{}")
+        except Exception as e:
+            return {"error": {"code": "INVALID_JSON", "message": f"Failed to parse JSON params: {e}"}}
+        return sanitize_keys(handle_run(params))
+
+    def get_task_status(task_id: str = "") -> dict:
+        """Query the progress, status, and final generated results of an async image generation task."""
+        return sanitize_keys(handle_get_task_status(task_id.strip()))
+
+    funcs = [
+        get_task_list,
+        get_model_architecture_list,
+        get_model_list,
+        get_feature_list,
+        get_model_features,
+        run,
+        get_task_status,
+    ]
+
+    for func in funcs:
+        gr.api(func)
+
+    for fn in demo.fns.values():
+        if getattr(fn, "api_name", None) in HIGH_LEVEL_MCP_API_NAMES:
+            fn.show_api = True
+
+    print("[MCP Integration] Successfully registered 7 High-Level Abstract MCP APIs via gr.api().")

mcp_tools/run.py

@@ -1,76 +1,76 @@
-"""
-MCP Tool: run
-Unified image generation task submission and execution interface.
-"""
-
-import time
-import uuid
-import threading
-from .common import (
-    _load_yaml,
-    _MODEL_LIST_PATH,
-    _TASK_DEFINITIONS,
-    _TASKS_DB,
-    _execute_imagegen_pipeline,
-)
-from .error_schema import make_validation_error, make_not_found_error
-
-
-def handle_run(params: dict) -> dict:
-    """Unified image generation task execution interface."""
-    if not isinstance(params, dict):
-        return make_validation_error("Request params must be an object.")
-
-    missing = []
-    for req_field in ["task_type", "model", "prompt"]:
-        if req_field not in params or not params[req_field]:
-            missing.append(req_field)
-    if missing:
-        return make_validation_error(
-            f"Missing required parameter(s): {', '.join(missing)}",
-            missing_fields=missing,
-        )
-
-    task_type = params["task_type"]
-    valid_tasks = [t["task_type"] for t in _TASK_DEFINITIONS]
-    if task_type not in valid_tasks:
-        return make_validation_error(
-            f"Invalid task_type '{task_type}'. Must be one of {valid_tasks}.",
-            invalid_fields={"task_type": f"Must be in {valid_tasks}"},
-        )
-
-    model_list = _load_yaml(_MODEL_LIST_PATH)
-    checkpoints = model_list.get("Checkpoint", {})
-    all_models = set()
-    for arch_name, arch_data in checkpoints.items():
-        if isinstance(arch_data, dict):
-            for m in arch_data.get("models", []):
-                all_models.add(m.get("display_name"))
-
-    if params["model"] not in all_models:
-        return make_not_found_error("model", params["model"])
-
-    task_id = f"img_task_{uuid.uuid4().hex[:10]}"
-    created_at = int(time.time())
-
-    _TASKS_DB[task_id] = {
-        "task_id": task_id,
-        "status": "queued",
-        "progress": 0,
-        "created_at": created_at,
-    }
-
-    async_exec = params.get("async_execution", False)
-
-    if async_exec:
-        t = threading.Thread(target=_execute_imagegen_pipeline, args=(task_id, params), daemon=True)
-        t.start()
-        return {
-            "status": "queued",
-            "task_id": task_id,
-            "poll_interval_ms": 2000,
-            "message": "Task queued successfully. Poll get_task_status for results.",
-        }
-    else:
-        _execute_imagegen_pipeline(task_id, params)
-        return _TASKS_DB[task_id]
+"""
+MCP Tool: run
+Unified image generation task submission and execution interface.
+"""
+
+import time
+import uuid
+import threading
+from .common import (
+    _load_yaml,
+    _MODEL_LIST_PATH,
+    _TASK_DEFINITIONS,
+    _TASKS_DB,
+    _execute_imagegen_pipeline,
+)
+from .error_schema import make_validation_error, make_not_found_error
+
+
+def handle_run(params: dict) -> dict:
+    """Unified image generation task execution interface."""
+    if not isinstance(params, dict):
+        return make_validation_error("Request params must be an object.")
+
+    missing = []
+    for req_field in ["task_type", "model", "prompt"]:
+        if req_field not in params or not params[req_field]:
+            missing.append(req_field)
+    if missing:
+        return make_validation_error(
+            f"Missing required parameter(s): {', '.join(missing)}",
+            missing_fields=missing,
+        )
+
+    task_type = params["task_type"]
+    valid_tasks = [t["task_type"] for t in _TASK_DEFINITIONS]
+    if task_type not in valid_tasks:
+        return make_validation_error(
+            f"Invalid task_type '{task_type}'. Must be one of {valid_tasks}.",
+            invalid_fields={"task_type": f"Must be in {valid_tasks}"},
+        )
+
+    model_list = _load_yaml(_MODEL_LIST_PATH)
+    checkpoints = model_list.get("Checkpoint", {})
+    all_models = set()
+    for arch_name, arch_data in checkpoints.items():
+        if isinstance(arch_data, dict):
+            for m in arch_data.get("models", []):
+                all_models.add(m.get("display_name"))
+
+    if params["model"] not in all_models:
+        return make_not_found_error("model", params["model"])
+
+    task_id = f"img_task_{uuid.uuid4().hex[:10]}"
+    created_at = int(time.time())
+
+    _TASKS_DB[task_id] = {
+        "task_id": task_id,
+        "status": "queued",
+        "progress": 0,
+        "created_at": created_at,
+    }
+
+    async_exec = params.get("async_execution", False)
+
+    if async_exec:
+        t = threading.Thread(target=_execute_imagegen_pipeline, args=(task_id, params), daemon=True)
+        t.start()
+        return {
+            "status": "queued",
+            "task_id": task_id,
+            "poll_interval_ms": 2000,
+            "message": "Task queued successfully. Poll get_task_status for results.",
+        }
+    else:
+        _execute_imagegen_pipeline(task_id, params)
+        return _TASKS_DB[task_id]

mcp_tools/tool_handlers.py

@@ -1,28 +1,28 @@
-"""
-MCP Tool Handlers — Backward-compatible aggregation entry point.
-Core logic has been split into individual files (get_*.py and run.py).
-"""
-
-from .get_task_list import handle_get_task_list
-from .get_model_architecture_list import handle_get_model_architecture_list
-from .get_model_list import handle_get_model_list
-from .get_feature_list import handle_get_feature_list
-from .get_model_features import handle_get_model_features
-from .run import handle_run
-from .get_task_status import handle_get_task_status
-from .common import (
-    _TASK_DEFINITIONS,
-    _TASKS_DB,
-    _load_yaml,
-    _execute_imagegen_pipeline,
-)
-
-__all__ = [
-    "handle_get_task_list",
-    "handle_get_model_architecture_list",
-    "handle_get_model_list",
-    "handle_get_feature_list",
-    "handle_get_model_features",
-    "handle_run",
-    "handle_get_task_status",
-]
+"""
+MCP Tool Handlers — Backward-compatible aggregation entry point.
+Core logic has been split into individual files (get_*.py and run.py).
+"""
+
+from .get_task_list import handle_get_task_list
+from .get_model_architecture_list import handle_get_model_architecture_list
+from .get_model_list import handle_get_model_list
+from .get_feature_list import handle_get_feature_list
+from .get_model_features import handle_get_model_features
+from .run import handle_run
+from .get_task_status import handle_get_task_status
+from .common import (
+    _TASK_DEFINITIONS,
+    _TASKS_DB,
+    _load_yaml,
+    _execute_imagegen_pipeline,
+)
+
+__all__ = [
+    "handle_get_task_list",
+    "handle_get_model_architecture_list",
+    "handle_get_model_list",
+    "handle_get_feature_list",
+    "handle_get_model_features",
+    "handle_run",
+    "handle_get_task_status",
+]

ui/layout.py

@@ -12,7 +12,7 @@ def build_ui(event_handler_function):
     with gr.Blocks() as demo:
         gr.Markdown("# ImageGen")
         gr.Markdown(
-            "This demo is a streamlined version of the [Comfy web UI](https://github.com/RioShiina47/comfy-webui)'s [ImageGen](https://huggingface.co/spaces/RioShiina/ImageGen) functionality."
+            "This demo is a streamlined version of the [Comfy web UI](https://github.com/RioShiina47/comfy-webui)'s [ImageGen](https://huggingface.co/spaces/RioShiina/ImageGen) functionality. Support [High-Level MCP](https://rioshiina-imagegen.hf.space/gradio_api/mcp/) 🤖"
         )
         with gr.Tabs(elem_id="tabs_container") as tabs:
             with gr.TabItem("Txt2Img", id=0):

yaml/chain_features.yaml

@@ -1,652 +1,652 @@
-# Complete Feature & Chain Definitions Configuration for MCP Tools
-# Every chain injector in chain_injectors/ corresponds 1-to-1 with an entry here (21 injectors total).
-
-lora:
-  chains: lora
-  display_name: "LoRA Fine-tuning Injector"
-  description: "Injects LoRA weights into UNet/DiT model and CLIP text encoder for custom style, character, or domain adaptation."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), then provide the lora_value (Civitai Version ID or HF repo path), and a single scale value (0.0~2.0) that controls both model and clip strength simultaneously."
-  parameters_schema:
-    type: object
-    properties:
-      source:
-        type: string
-        enum: ["Civitai", "Hugging Face"]
-        description: "Download source for the LoRA model. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo path."
-      lora_value:
-        type: string
-        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456). For Hugging Face: repo_id/filename.extension or repo_id/folder_path/filename.extension (e.g., 'lightx2v/Qwen-Image-Lightning/Qwen-Image-Lightning-4steps-V2.0-bf16.safetensors')."
-      scale:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Unified strength applied to both the UNet/DiT model and CLIP text encoder (0.0 to 2.0)."
-    required:
-      - source
-      - lora_value
-
-ipadapter:
-  chains: ipadapter
-  display_name: "IP-Adapter Image Prompt"
-  description: "Uses reference images to guide generation style, composition, structure, or face appearance without prompt text restrictions (SD1.5 & SDXL)."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply global settings (preset, embeds_scaling, combine_method, final_weight) and up to 5 reference images with individual weights. Preset must match the target model architecture (SD1.5 or SDXL)."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI (e.g., data:image/png;base64,...) or HTTP/HTTPS URL."
-      weight:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Influence weight of the individual image prompt (0.0 to 2.0)."
-      preset:
-        type: string
-        default: "STANDARD (medium strength)"
-        description: "IPAdapter preset model variant loaded from ipadapter.yaml. Must match model architecture (SD1.5 vs SDXL)."
-      embeds_scaling:
-        type: string
-        default: "V only"
-        enum:
-          - "V only"
-          - "K+V"
-          - "K+V w/ C penalty"
-          - "K+mean(V) w/ C penalty"
-        description: "Embedding scaling method for IPAdapter."
-      combine_method:
-        type: string
-        default: "concat"
-        enum:
-          - "concat"
-          - "add"
-          - "subtract"
-          - "average"
-          - "norm average"
-          - "max"
-          - "min"
-        description: "Combination method for multiple reference images."
-      final_weight:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Global weight multiplier for IPAdapter conditioning (0.0 to 2.0)."
-      lora_strength:
-        type: number
-        default: 0.6
-        description: "LoRA weight strength for FaceID adapter variants."
-    required:
-      - image
-
-controlnet:
-  chains: controlnet
-  display_name: "ControlNet Spatial Guidance"
-  description: "Applies structural and spatial conditioning (depth, pose, lineart, tile, scribble, canny) to guide output composition."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Specify ControlNet type and series (must match requested model architecture e.g., SD1.5, SDXL, SD3.5, FLUX.1, Qwen-Image), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and guidance strength."
-  parameters_schema:
-    type: object
-    properties:
-      type:
-        type: string
-        description: "ControlNet conditioning type (must match model architecture)."
-      series:
-        type: string
-        description: "ControlNet model series (must match model architecture)."
-      image:
-        type: string
-        description: "Pre-processed control image (e.g., Depth, Canny, Pose, Lineart map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
-      strength:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Control influence strength (0.0 to 2.0)."
-    required:
-      - type
-      - series
-      - image
-
-conditioning:
-  chains: conditioning
-  display_name: "Regional Conditioning / Area Prompt"
-  description: "Defines rectangular areas (X, Y, Width, Height) and assigns specific text prompts and conditioning strengths to them."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 10
-  usage_guideline: "Define rectangular spatial areas (X, Y, width, height) and assign specific prompts and strengths to them. Supports up to 10 area prompts."
-  parameters_schema:
-    type: object
-    properties:
-      prompt:
-        type: string
-        description: "Text prompt for this specific rectangular area."
-      x:
-        type: integer
-        default: 0
-        description: "Top-left X coordinate of the rectangular area."
-      y:
-        type: integer
-        default: 0
-        description: "Top-left Y coordinate of the rectangular area."
-      width:
-        type: integer
-        default: 512
-        description: "Width of the rectangular area."
-      height:
-        type: integer
-        default: 512
-        description: "Height of the rectangular area."
-      strength:
-        type: number
-        default: 1.0
-        minimum: 0.1
-        maximum: 2.0
-        description: "Conditioning strength for this area (0.1 to 2.0)."
-    required:
-      - prompt
-
-vae:
-  chains: vae
-  display_name: "Custom VAE Loader"
-  description: "Overrides default VAE model used for latent space encoding and final image decoding."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 1
-  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), then provide the vae_value (Civitai Version ID or HF file path)."
-  parameters_schema:
-    type: object
-    properties:
-      source:
-        type: string
-        enum: ["Civitai", "Hugging Face"]
-        description: "Download source for the VAE model. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo path."
-      vae_value:
-        type: string
-        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456). For Hugging Face: repo_id/filename.extension or repo_id/folder_path/filename.extension (e.g., 'madebyollin/sdxl-vae-fp16-fix/sdxl_vae.safetensors')."
-    required:
-      - source
-      - vae_value
-
-pid:
-  chains: pid
-  display_name: "PiD High-Resolution Refinement"
-  description: "Progressive Detail (PiD) upscale injector for fine detail enhancement and resolution upscaling."
-  supported_tasks:
-    - txt2img
-  max_count: 1
-  usage_guideline: "Enables PiD detail refinement pipeline using a boolean switch ('enabled': true/false)."
-  parameters_schema:
-    type: object
-    properties:
-      enabled:
-        type: boolean
-        default: true
-        description: "Enable or disable PiD High-Resolution Refinement."
-    required:
-      - enabled
-
-flux1_style:
-  chains: style
-  display_name: "FLUX.1 Style Reference"
-  description: "Applies artistic style conditioning from reference images onto FLUX.1 model generated outputs."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply style reference image(s) (up to 5) encoded as Base64 Data URI or HTTP/HTTPS URL and optional strength."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Style reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-      strength:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Style influence strength (0.0 to 2.0)."
-    required:
-      - image
-
-reference_edit:
-  chains: reference_latent
-  display_name: "Reference Edit"
-  description: "For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 10
-  usage_guideline: "Supply reference image(s) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images (up to 10) performs an Image Combine."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-mage_flow_reference_edit:
-  chains: reference_image
-  display_name: "Mage-Flow Reference Edit"
-  description: " (Mage-Flow-Edit-Turbo/Mage-Flow-Edit recommended) For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 10
-  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-krea2_identity_edit:
-  chains: krea2_identity_edit
-  display_name: "KREA2 Identity Edit"
-  description: "Processed using the lbouaraba/comfyui-krea2edit node. (Krea-2-Turbo recommended, Krea-2-Raw need set ZeroGPU Duration (s) to 120 ) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 2
-  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-krea2_style_reference:
-  chains: krea2_style_reference
-  display_name: "KREA2 Style Reference"
-  description: "(Krea-2-Turbo recommended) Add style reference images to perform style reference editing."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 3
-  usage_guideline: "Supply style reference image as Base64 Data URI or HTTP/HTTPS URL."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Style reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-diffsynth_controlnet:
-  chains: diffsynth_controlnet
-  display_name: "DiffSynth ControlNet"
-  description: "DiffSynth optimized ControlNet injector for Z-Image models."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply ControlNet type (e.g., 'Canny'), series (e.g., 'alibaba-pai Controlnet Union 2.1 8steps'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
-  parameters_schema:
-    type: object
-    properties:
-      type:
-        type: string
-        description: "ControlNet conditioning type."
-      series:
-        type: string
-        description: "ControlNet model series name."
-      image:
-        type: string
-        description: "Pre-processed control image (e.g., Depth map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
-      strength:
-        type: number
-        default: 1.0
-        description: "Control influence strength."
-    required:
-      - type
-      - series
-      - image
-
-boogu_image_edit:
-  chains: boogu_image_edit
-  display_name: "Boogu-Image Edit"
-  description: " (Boogu-Image-Edit-Turbo/Boogu-Image-Edit recommended, Boogu-Image-Edit need set ZeroGPU Duration (s) to 120 ) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 2
-  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-joyai_reference_edit:
-  chains: joyai_image
-  display_name: "JoyAI Reference Edit"
-  description: " (JoyAI-Image-Edit recommended) For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit (JoyAI-Image-Edit recommended), while adding multiple images performs an Image Combine (JoyAI-Image-Edit-Plus recommended with ZeroGPU Duration (s) set to 120)."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 2
-  usage_guideline: "Supply JoyAI reference image as Base64 Data URI or HTTP/HTTPS URL."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Input reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-qwen_image_edit:
-  chains: qwen_image_edit
-  display_name: "Qwen-Image Edit"
-  description: " (lightx2v/Qwen-Image-Edit-2511-Lightning recommended) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 3
-  usage_guideline: "Supply reference image(s) (up to 3) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images performs an Image Combine."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-hidream_o1_smoothing:
-  chains: hidream_o1_smoothing
-  display_name: "HiDream O1 Smoothing Injector"
-  description: "HiDream O1 detail smoothing and artifact reduction injector."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 1
-  usage_guideline: "Configures smoothing factor for HiDream models."
-  parameters_schema:
-    type: object
-    properties:
-      factor:
-        type: number
-        default: 0.5
-        description: "Smoothing intensity (0.0 to 1.0)."
-    required: []
-
-krea2_controlnet:
-  chains: krea2_controlnet
-  display_name: "KREA2 ControlNet"
-  description: "Processed using the facok/comfyui-krea2-controlnet node."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply ControlNet type (e.g., 'Depth'), series (e.g., 'Patil'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
-  parameters_schema:
-    type: object
-    properties:
-      type:
-        type: string
-        enum:
-          - "Depth"
-        description: "ControlNet conditioning type."
-      series:
-        type: string
-        enum:
-          - "Patil"
-        default: "Patil"
-        description: "ControlNet model series."
-      image:
-        type: string
-        description: "Pre-processed control image (e.g., Depth map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
-      strength:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Control influence strength (0.0 to 2.0)."
-    required:
-      - type
-      - series
-      - image
-
-anima_controlnet_lllite:
-  chains: anima_controlnet_lllite
-  display_name: "Anima ControlNet LLLite"
-  description: "Anima model-specific lightweight ControlNet."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply Anima ControlNet LLLite type (e.g., 'Depth'), series (e.g., 'kohya-ss'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
-  parameters_schema:
-    type: object
-    properties:
-      type:
-        type: string
-        description: "Anima ControlNet LLLite conditioning type."
-      series:
-        type: string
-        description: "Anima ControlNet LLLite model series."
-      image:
-        type: string
-        description: "Pre-processed control image (e.g., Depth, Lineart map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
-      strength:
-        type: number
-        default: 1.0
-        minimum: 0.0
-        maximum: 2.0
-        description: "Control influence strength (0.0 to 2.0)."
-    required:
-      - type
-      - series
-      - image
-
-hidream_o1_reference:
-  chains: hidream_o1_reference
-  display_name: "HiDream-O1 Reference Edit"
-  description: " (HiDream-O1-Image-Dev recommended with resolution set to 4.0MP, e.g., 2048x2048) For HiDream-O1 models, this feature enables reference image editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 9
-  usage_guideline: "Supply reference image(s) (up to 9) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images performs an Image Combine."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-    required:
-      - image
-
-flux1_ipadapter:
-  chains: flux1_ipadapter
-  display_name: "Flux1 IP-Adapter"
-  description: "FLUX.1 model-specific IP-Adapter implementation."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply reference image (Base64 Data URI or HTTP/HTTPS URL), optional weight (default 1.0), start_at (default 0.0), and end_at (default 1.0). Up to 5 images supported."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-      weight:
-        type: number
-        default: 1.0
-        description: "Influence weight of the image prompt (0.0 to 2.0)."
-      start_at:
-        type: number
-        default: 0.0
-        description: "Start step percentage for IP-Adapter application (0.0 to 1.0)."
-      end_at:
-        type: number
-        default: 1.0
-        description: "End step percentage for IP-Adapter application (0.0 to 1.0)."
-      start_percent:
-        type: number
-        default: 0.0
-        description: "Alias for start_at."
-      end_percent:
-        type: number
-        default: 1.0
-        description: "Alias for end_at."
-    required:
-      - image
-
-sd3_ipadapter:
-  chains: sd3_ipadapter
-  display_name: "SD3 IP-Adapter"
-  description: "SD3/SD3.5 model-specific IP-Adapter implementation."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Supply reference image (Base64 Data URI or HTTP/HTTPS URL), optional weight (default 1.0), start_at (default 0.0), and end_at (default 1.0). Up to 5 images supported."
-  parameters_schema:
-    type: object
-    properties:
-      image:
-        type: string
-        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
-      weight:
-        type: number
-        default: 1.0
-        description: "Influence weight of the image prompt (0.0 to 2.0)."
-      start_at:
-        type: number
-        default: 0.0
-        description: "Start step percentage for IP-Adapter application (0.0 to 1.0)."
-      end_at:
-        type: number
-        default: 1.0
-        description: "End step percentage for IP-Adapter application (0.0 to 1.0)."
-      start_percent:
-        type: number
-        default: 0.0
-        description: "Alias for start_at."
-      end_percent:
-        type: number
-        default: 1.0
-        description: "Alias for end_at."
-    required:
-      - image
-
-embedding:
-  chains: embedding
-  display_name: "Textual Inversion Embedding Injector"
-  description: "Downloads Textual Inversion embedding files from Civitai or Hugging Face to the server. Note: This feature ONLY handles file downloading/preparation. To activate the embedding, manually add 'embedding:<filename>' (e.g. 'embedding:civitai_456' for Civitai ID 456, or 'embedding:filename' for Hugging Face) into your prompt or negative_prompt."
-  supported_tasks:
-    - txt2img
-    - img2img
-    - inpaint
-    - outpaint
-    - hires_fix
-  max_count: 5
-  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), and embedding_value (Civitai Version ID or HF repo file path). The file is downloaded to server; manually enter 'embedding:<filename>' in prompt or negative_prompt to activate."
-  parameters_schema:
-    type: object
-    properties:
-      source:
-        type: string
-        enum:
-          - "Civitai"
-          - "Hugging Face"
-        description: "Download source for the Textual Inversion embedding file. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo file path."
-      embedding_value:
-        type: string
-        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456, saved as 'civitai_456.safetensors'). For Hugging Face: repo_id/filename.extension (e.g., 'ilikebigturtles/lazypos/lazypos.safetensors', saved as 'lazypos.safetensors'). Manually reference embedding:<filename> in prompt or negative_prompt."
-    required:
-      - source
-      - embedding_value
+# Complete Feature & Chain Definitions Configuration for MCP Tools
+# Every chain injector in chain_injectors/ corresponds 1-to-1 with an entry here (21 injectors total).
+
+lora:
+  chains: lora
+  display_name: "LoRA Fine-tuning Injector"
+  description: "Injects LoRA weights into UNet/DiT model and CLIP text encoder for custom style, character, or domain adaptation."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), then provide the lora_value (Civitai Version ID or HF repo path), and a single scale value (0.0~2.0) that controls both model and clip strength simultaneously."
+  parameters_schema:
+    type: object
+    properties:
+      source:
+        type: string
+        enum: ["Civitai", "Hugging Face"]
+        description: "Download source for the LoRA model. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo path."
+      lora_value:
+        type: string
+        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456). For Hugging Face: repo_id/filename.extension or repo_id/folder_path/filename.extension (e.g., 'lightx2v/Qwen-Image-Lightning/Qwen-Image-Lightning-4steps-V2.0-bf16.safetensors')."
+      scale:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Unified strength applied to both the UNet/DiT model and CLIP text encoder (0.0 to 2.0)."
+    required:
+      - source
+      - lora_value
+
+ipadapter:
+  chains: ipadapter
+  display_name: "IP-Adapter Image Prompt"
+  description: "Uses reference images to guide generation style, composition, structure, or face appearance without prompt text restrictions (SD1.5 & SDXL)."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply global settings (preset, embeds_scaling, combine_method, final_weight) and up to 5 reference images with individual weights. Preset must match the target model architecture (SD1.5 or SDXL)."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI (e.g., data:image/png;base64,...) or HTTP/HTTPS URL."
+      weight:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Influence weight of the individual image prompt (0.0 to 2.0)."
+      preset:
+        type: string
+        default: "STANDARD (medium strength)"
+        description: "IPAdapter preset model variant loaded from ipadapter.yaml. Must match model architecture (SD1.5 vs SDXL)."
+      embeds_scaling:
+        type: string
+        default: "V only"
+        enum:
+          - "V only"
+          - "K+V"
+          - "K+V w/ C penalty"
+          - "K+mean(V) w/ C penalty"
+        description: "Embedding scaling method for IPAdapter."
+      combine_method:
+        type: string
+        default: "concat"
+        enum:
+          - "concat"
+          - "add"
+          - "subtract"
+          - "average"
+          - "norm average"
+          - "max"
+          - "min"
+        description: "Combination method for multiple reference images."
+      final_weight:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Global weight multiplier for IPAdapter conditioning (0.0 to 2.0)."
+      lora_strength:
+        type: number
+        default: 0.6
+        description: "LoRA weight strength for FaceID adapter variants."
+    required:
+      - image
+
+controlnet:
+  chains: controlnet
+  display_name: "ControlNet Spatial Guidance"
+  description: "Applies structural and spatial conditioning (depth, pose, lineart, tile, scribble, canny) to guide output composition."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Specify ControlNet type and series (must match requested model architecture e.g., SD1.5, SDXL, SD3.5, FLUX.1, Qwen-Image), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and guidance strength."
+  parameters_schema:
+    type: object
+    properties:
+      type:
+        type: string
+        description: "ControlNet conditioning type (must match model architecture)."
+      series:
+        type: string
+        description: "ControlNet model series (must match model architecture)."
+      image:
+        type: string
+        description: "Pre-processed control image (e.g., Depth, Canny, Pose, Lineart map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
+      strength:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Control influence strength (0.0 to 2.0)."
+    required:
+      - type
+      - series
+      - image
+
+conditioning:
+  chains: conditioning
+  display_name: "Regional Conditioning / Area Prompt"
+  description: "Defines rectangular areas (X, Y, Width, Height) and assigns specific text prompts and conditioning strengths to them."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 10
+  usage_guideline: "Define rectangular spatial areas (X, Y, width, height) and assign specific prompts and strengths to them. Supports up to 10 area prompts."
+  parameters_schema:
+    type: object
+    properties:
+      prompt:
+        type: string
+        description: "Text prompt for this specific rectangular area."
+      x:
+        type: integer
+        default: 0
+        description: "Top-left X coordinate of the rectangular area."
+      y:
+        type: integer
+        default: 0
+        description: "Top-left Y coordinate of the rectangular area."
+      width:
+        type: integer
+        default: 512
+        description: "Width of the rectangular area."
+      height:
+        type: integer
+        default: 512
+        description: "Height of the rectangular area."
+      strength:
+        type: number
+        default: 1.0
+        minimum: 0.1
+        maximum: 2.0
+        description: "Conditioning strength for this area (0.1 to 2.0)."
+    required:
+      - prompt
+
+vae:
+  chains: vae
+  display_name: "Custom VAE Loader"
+  description: "Overrides default VAE model used for latent space encoding and final image decoding."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 1
+  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), then provide the vae_value (Civitai Version ID or HF file path)."
+  parameters_schema:
+    type: object
+    properties:
+      source:
+        type: string
+        enum: ["Civitai", "Hugging Face"]
+        description: "Download source for the VAE model. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo path."
+      vae_value:
+        type: string
+        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456). For Hugging Face: repo_id/filename.extension or repo_id/folder_path/filename.extension (e.g., 'madebyollin/sdxl-vae-fp16-fix/sdxl_vae.safetensors')."
+    required:
+      - source
+      - vae_value
+
+pid:
+  chains: pid
+  display_name: "PiD High-Resolution Refinement"
+  description: "Progressive Detail (PiD) upscale injector for fine detail enhancement and resolution upscaling."
+  supported_tasks:
+    - txt2img
+  max_count: 1
+  usage_guideline: "Enables PiD detail refinement pipeline using a boolean switch ('enabled': true/false)."
+  parameters_schema:
+    type: object
+    properties:
+      enabled:
+        type: boolean
+        default: true
+        description: "Enable or disable PiD High-Resolution Refinement."
+    required:
+      - enabled
+
+flux1_style:
+  chains: style
+  display_name: "FLUX.1 Style Reference"
+  description: "Applies artistic style conditioning from reference images onto FLUX.1 model generated outputs."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply style reference image(s) (up to 5) encoded as Base64 Data URI or HTTP/HTTPS URL and optional strength."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Style reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+      strength:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Style influence strength (0.0 to 2.0)."
+    required:
+      - image
+
+reference_edit:
+  chains: reference_latent
+  display_name: "Reference Edit"
+  description: "For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 10
+  usage_guideline: "Supply reference image(s) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images (up to 10) performs an Image Combine."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+mage_flow_reference_edit:
+  chains: reference_image
+  display_name: "Mage-Flow Reference Edit"
+  description: " (Mage-Flow-Edit-Turbo/Mage-Flow-Edit recommended) For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 10
+  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+krea2_identity_edit:
+  chains: krea2_identity_edit
+  display_name: "KREA2 Identity Edit"
+  description: "Processed using the lbouaraba/comfyui-krea2edit node. (Krea-2-Turbo recommended, Krea-2-Raw need set ZeroGPU Duration (s) to 120 ) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 2
+  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+krea2_style_reference:
+  chains: krea2_style_reference
+  display_name: "KREA2 Style Reference"
+  description: "(Krea-2-Turbo recommended) Add style reference images to perform style reference editing."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 3
+  usage_guideline: "Supply style reference image as Base64 Data URI or HTTP/HTTPS URL."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Style reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+diffsynth_controlnet:
+  chains: diffsynth_controlnet
+  display_name: "DiffSynth ControlNet"
+  description: "DiffSynth optimized ControlNet injector for Z-Image models."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply ControlNet type (e.g., 'Canny'), series (e.g., 'alibaba-pai Controlnet Union 2.1 8steps'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
+  parameters_schema:
+    type: object
+    properties:
+      type:
+        type: string
+        description: "ControlNet conditioning type."
+      series:
+        type: string
+        description: "ControlNet model series name."
+      image:
+        type: string
+        description: "Pre-processed control image (e.g., Depth map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
+      strength:
+        type: number
+        default: 1.0
+        description: "Control influence strength."
+    required:
+      - type
+      - series
+      - image
+
+boogu_image_edit:
+  chains: boogu_image_edit
+  display_name: "Boogu-Image Edit"
+  description: " (Boogu-Image-Edit-Turbo/Boogu-Image-Edit recommended, Boogu-Image-Edit need set ZeroGPU Duration (s) to 120 ) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 2
+  usage_guideline: "Supply reference image as Base64 Data URI or HTTP/HTTPS URL."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+joyai_reference_edit:
+  chains: joyai_image
+  display_name: "JoyAI Reference Edit"
+  description: " (JoyAI-Image-Edit recommended) For multimodal models, this feature enables powerful editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit (JoyAI-Image-Edit recommended), while adding multiple images performs an Image Combine (JoyAI-Image-Edit-Plus recommended with ZeroGPU Duration (s) set to 120)."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 2
+  usage_guideline: "Supply JoyAI reference image as Base64 Data URI or HTTP/HTTPS URL."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Input reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+qwen_image_edit:
+  chains: qwen_image_edit
+  display_name: "Qwen-Image Edit"
+  description: " (lightx2v/Qwen-Image-Edit-2511-Lightning recommended) In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 3
+  usage_guideline: "Supply reference image(s) (up to 3) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images performs an Image Combine."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+hidream_o1_smoothing:
+  chains: hidream_o1_smoothing
+  display_name: "HiDream O1 Smoothing Injector"
+  description: "HiDream O1 detail smoothing and artifact reduction injector."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 1
+  usage_guideline: "Configures smoothing factor for HiDream models."
+  parameters_schema:
+    type: object
+    properties:
+      factor:
+        type: number
+        default: 0.5
+        description: "Smoothing intensity (0.0 to 1.0)."
+    required: []
+
+krea2_controlnet:
+  chains: krea2_controlnet
+  display_name: "KREA2 ControlNet"
+  description: "Processed using the facok/comfyui-krea2-controlnet node."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply ControlNet type (e.g., 'Depth'), series (e.g., 'Patil'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
+  parameters_schema:
+    type: object
+    properties:
+      type:
+        type: string
+        enum:
+          - "Depth"
+        description: "ControlNet conditioning type."
+      series:
+        type: string
+        enum:
+          - "Patil"
+        default: "Patil"
+        description: "ControlNet model series."
+      image:
+        type: string
+        description: "Pre-processed control image (e.g., Depth map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
+      strength:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Control influence strength (0.0 to 2.0)."
+    required:
+      - type
+      - series
+      - image
+
+anima_controlnet_lllite:
+  chains: anima_controlnet_lllite
+  display_name: "Anima ControlNet LLLite"
+  description: "Anima model-specific lightweight ControlNet."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply Anima ControlNet LLLite type (e.g., 'Depth'), series (e.g., 'kohya-ss'), pre-processed control image (Base64 Data URI or HTTP/HTTPS URL; system does NOT pre-process raw RGB images), and optional strength."
+  parameters_schema:
+    type: object
+    properties:
+      type:
+        type: string
+        description: "Anima ControlNet LLLite conditioning type."
+      series:
+        type: string
+        description: "Anima ControlNet LLLite model series."
+      image:
+        type: string
+        description: "Pre-processed control image (e.g., Depth, Lineart map) encoded as Base64 Data URI or HTTP/HTTPS URL. Note: System does NOT automatically pre-process raw RGB images."
+      strength:
+        type: number
+        default: 1.0
+        minimum: 0.0
+        maximum: 2.0
+        description: "Control influence strength (0.0 to 2.0)."
+    required:
+      - type
+      - series
+      - image
+
+hidream_o1_reference:
+  chains: hidream_o1_reference
+  display_name: "HiDream-O1 Reference Edit"
+  description: " (HiDream-O1-Image-Dev recommended with resolution set to 4.0MP, e.g., 2048x2048) For HiDream-O1 models, this feature enables reference image editing and combining capabilities. In txt2img mode, adding a single reference image performs an Image Edit, while adding multiple images performs an Image Combine."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 9
+  usage_guideline: "Supply reference image(s) (up to 9) encoded as Base64 Data URI or HTTP/HTTPS URL. Passing a single reference image performs an Image Edit, while passing multiple images performs an Image Combine."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+    required:
+      - image
+
+flux1_ipadapter:
+  chains: flux1_ipadapter
+  display_name: "Flux1 IP-Adapter"
+  description: "FLUX.1 model-specific IP-Adapter implementation."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply reference image (Base64 Data URI or HTTP/HTTPS URL), optional weight (default 1.0), start_at (default 0.0), and end_at (default 1.0). Up to 5 images supported."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+      weight:
+        type: number
+        default: 1.0
+        description: "Influence weight of the image prompt (0.0 to 2.0)."
+      start_at:
+        type: number
+        default: 0.0
+        description: "Start step percentage for IP-Adapter application (0.0 to 1.0)."
+      end_at:
+        type: number
+        default: 1.0
+        description: "End step percentage for IP-Adapter application (0.0 to 1.0)."
+      start_percent:
+        type: number
+        default: 0.0
+        description: "Alias for start_at."
+      end_percent:
+        type: number
+        default: 1.0
+        description: "Alias for end_at."
+    required:
+      - image
+
+sd3_ipadapter:
+  chains: sd3_ipadapter
+  display_name: "SD3 IP-Adapter"
+  description: "SD3/SD3.5 model-specific IP-Adapter implementation."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Supply reference image (Base64 Data URI or HTTP/HTTPS URL), optional weight (default 1.0), start_at (default 0.0), and end_at (default 1.0). Up to 5 images supported."
+  parameters_schema:
+    type: object
+    properties:
+      image:
+        type: string
+        description: "Reference image encoded as Base64 Data URI or HTTP/HTTPS URL."
+      weight:
+        type: number
+        default: 1.0
+        description: "Influence weight of the image prompt (0.0 to 2.0)."
+      start_at:
+        type: number
+        default: 0.0
+        description: "Start step percentage for IP-Adapter application (0.0 to 1.0)."
+      end_at:
+        type: number
+        default: 1.0
+        description: "End step percentage for IP-Adapter application (0.0 to 1.0)."
+      start_percent:
+        type: number
+        default: 0.0
+        description: "Alias for start_at."
+      end_percent:
+        type: number
+        default: 1.0
+        description: "Alias for end_at."
+    required:
+      - image
+
+embedding:
+  chains: embedding
+  display_name: "Textual Inversion Embedding Injector"
+  description: "Downloads Textual Inversion embedding files from Civitai or Hugging Face to the server. Note: This feature ONLY handles file downloading/preparation. To activate the embedding, manually add 'embedding:<filename>' (e.g. 'embedding:civitai_456' for Civitai ID 456, or 'embedding:filename' for Hugging Face) into your prompt or negative_prompt."
+  supported_tasks:
+    - txt2img
+    - img2img
+    - inpaint
+    - outpaint
+    - hires_fix
+  max_count: 5
+  usage_guideline: "Specify source ('Civitai' or 'Hugging Face'), and embedding_value (Civitai Version ID or HF repo file path). The file is downloaded to server; manually enter 'embedding:<filename>' in prompt or negative_prompt to activate."
+  parameters_schema:
+    type: object
+    properties:
+      source:
+        type: string
+        enum:
+          - "Civitai"
+          - "Hugging Face"
+        description: "Download source for the Textual Inversion embedding file. Use 'Civitai' to download by Version ID, or 'Hugging Face' to download by repo file path."
+      embedding_value:
+        type: string
+        description: "For Civitai: the Version ID (e.g., '456' from civitai.com/models/123?modelVersionId=456, saved as 'civitai_456.safetensors'). For Hugging Face: repo_id/filename.extension (e.g., 'ilikebigturtles/lazypos/lazypos.safetensors', saved as 'lazypos.safetensors'). Manually reference embedding:<filename> in prompt or negative_prompt."
+    required:
+      - source
+      - embedding_value