scmcp-shared 0.3.6__py3-none-any.whl → 0.4.0__py3-none-any.whl

scmcp_shared/__init__.py

@@ -1,3 +1,3 @@
 
-__version__ = "0.3.6"
+__version__ = "0.4.0"
 

scmcp_shared/agent.py

@@ -0,0 +1,30 @@
+import instructor
+from openai import OpenAI
+from scmcp_shared.schema.tool import ToolList
+import os
+
+
+
+def select_tool(query):
+
+    API_KEY = os.environ.get("API_KEY",  None)
+    BASE_URL = os.environ.get("BASE_URL", None)
+    MODEL = os.environ.get("MODEL", None)
+
+    client = OpenAI(api_key=API_KEY, base_url=BASE_URL)
+    client = instructor.from_openai(client)
+
+    response = client.chat.completions.create(
+        model=MODEL,
+        messages=[
+            {
+                "role": "system", 
+                "content": f"you are a bioinformatician, you are given a task and a list of tools, you need to select the most directly relevant tools to use to solve the task"},
+            {
+                "role": "user",
+                "content": query
+            },  
+        ],
+        response_model=ToolList,
+    )
+    return response.tools

scmcp_shared/cli.py

@@ -35,6 +35,7 @@ class MCPCLI:
         parser.add_argument('--host', default='127.0.0.1', help='transport host')
         parser.add_argument('-f', '--forward', help='forward request to another server')
         parser.add_argument('-wd', '--working-dir', default=".", help='working directory')
+        parser.add_argument('--tool-mode', choices=["auto", "normal"], default="normal", help='tool selection mode')
         parser.add_argument('--log-file', help='log file path, use stdout if None')
     
     def add_command(self, name: str, help_text: str, handler: Callable) -> argparse.ArgumentParser:
@@ -78,6 +79,15 @@ class MCPCLI:
             modules = None
         if self.manager is not None:
             self.mcp = self.manager(self.name, include_modules=modules).mcp
+            all_tools = self.mcp._tool_manager._tools
+            auto_tools = {tool: all_tools[tool] for tool in all_tools if all_tools[tool].name in ["search_tool", "run_tool"]}
+            if args.tool_mode == "auto":
+                all_tools = self.mcp._tool_manager._tools
+                self.mcp._tool_manager._all_tools = all_tools
+                self.mcp._tool_manager._tools = auto_tools
+            else:
+                for name in auto_tools:
+                    self.mcp._tool_manager.remove_tool(name)
         elif self.mcp is not None:
             pass
         else:

scmcp_shared/schema/io.py

@@ -7,7 +7,7 @@ from typing import Optional, Literal
 
 
 
-class ReadModel(BaseModel):
+class ReadParams(BaseModel):
     """Input schema for the read tool."""
     filename: str = Field(
         ...,
@@ -82,7 +82,7 @@ class ReadModel(BaseModel):
         return v
 
 
-class WriteModel(BaseModel):
+class WriteParams(BaseModel):
     """Input schema for the write tool."""
     filename: str = Field(
         description="Path to save the file. If no extension is provided, the default format will be used."

scmcp_shared/schema/pl.py

@@ -72,12 +72,12 @@ class FigureSizeMixin:
 
 
 # 基础可视化模型，包含所有可视化工具共享的字段
-class BaseVisualizationModel(BaseModel, LegendMixin, ColorMappingMixin, FigureSizeMixin):
+class BaseVisualizationParams(BaseModel, LegendMixin, ColorMappingMixin, FigureSizeMixin):
     """基础可视化模型，包含所有可视化工具共享的字段"""    
     pass
 
 # 基础嵌入可视化模型，包含所有嵌入可视化工具共享的字段
-class BaseEmbeddingModel(BaseVisualizationModel):
+class BaseEmbeddingParams(BaseVisualizationParams):
     """基础嵌入可视化模型，包含所有嵌入可视化工具共享的字段"""
     
     color: Optional[Union[str, List[str]]] = Field(
@@ -176,8 +176,8 @@ class BaseEmbeddingModel(BaseVisualizationModel):
     )
 
 
-# 重构 ScatterModel 作为基础散点图模型
-class BaseScatterModel(BaseVisualizationModel):
+# 重构 ScatterParams 作为基础散点图模型
+class BaseScatterParams(BaseVisualizationParams):
     """基础散点图模型"""
     
     x: Optional[str] = Field(
@@ -211,8 +211,8 @@ class BaseScatterModel(BaseVisualizationModel):
     )
 
 
-# 使用继承关系重构 EnhancedScatterModel
-class EnhancedScatterModel(BaseScatterModel):
+# 使用继承关系重构 EnhancedScatterParams
+class EnhancedScatterParams(BaseScatterParams):
     """Input schema for the enhanced scatter plotting tool."""
     
     sort_order: bool = Field(
@@ -261,7 +261,7 @@ class EnhancedScatterModel(BaseScatterModel):
 
 
 # 创建基础统计可视化模型
-class BaseStatPlotModel(BaseVisualizationModel):
+class BaseStatPlotParams(BaseVisualizationParams):
     """基础统计可视化模型，包含统计图表共享的字段"""
     
     groupby: Optional[str] = Field(
@@ -328,8 +328,8 @@ class BaseStatPlotModel(BaseVisualizationModel):
             raise ValueError("size must be a positive integer")
         return v
 
-# 添加缺失的 BaseMatrixModel 类
-class BaseMatrixModel(BaseVisualizationModel):
+# 添加缺失的 BaseMatrixParams 类
+class BaseMatrixParams(BaseVisualizationParams):
     """基础矩阵可视化模型，包含所有矩阵可视化工具共享的字段"""
     
     var_names: Union[List[str], Mapping[str, List[str]]] = Field(
@@ -374,8 +374,8 @@ class BaseMatrixModel(BaseVisualizationModel):
     )
 
 
-# 重构 HeatmapModel
-class HeatmapModel(BaseMatrixModel):
+# 重构 HeatmapParams
+class HeatmapParams(BaseMatrixParams):
     """Input schema for the heatmap plotting tool."""
     
     num_categories: int = Field(
@@ -412,14 +412,14 @@ class HeatmapModel(BaseMatrixModel):
         return v
 
 
-# 重构 TracksplotModel
-class TracksplotModel(BaseMatrixModel):
+# 重构 TracksplotParams
+class TracksplotParams(BaseMatrixParams):
     """Input schema for the tracksplot plotting tool."""
-    # 所有需要的字段已经在 BaseMatrixModel 中定义
+    # 所有需要的字段已经在 BaseMatrixParams 中定义
 
 
-# 重构 ViolinModel
-class ViolinModel(BaseStatPlotModel):
+# 重构 ViolinParams
+class ViolinParams(BaseStatPlotParams):
     """Input schema for the violin plotting tool."""
     
     keys: Union[str, List[str]] = Field(
@@ -477,8 +477,8 @@ class ViolinModel(BaseStatPlotModel):
         return v
 
 
-# 重构 MatrixplotModel
-class MatrixplotModel(BaseMatrixModel):
+# 重构 MatrixplotParams
+class MatrixplotParams(BaseMatrixParams):
     """Input schema for the matrixplot plotting tool."""
     
     num_categories: int = Field(
@@ -523,8 +523,8 @@ class MatrixplotModel(BaseMatrixModel):
         return v
 
 
-# 重构 DotplotModel
-class DotplotModel(BaseMatrixModel):
+# 重构 DotplotParams
+class DotplotParams(BaseMatrixParams):
     """Input schema for the dotplot plotting tool."""
     
     expression_cutoff: float = Field(
@@ -577,8 +577,8 @@ class DotplotModel(BaseMatrixModel):
     )
 
 
-# 重构 RankGenesGroupsModel
-class RankGenesGroupsModel(BaseVisualizationModel):
+# 重构 RankGenesGroupsParams
+class RankGenesGroupsParams(BaseVisualizationParams):
     """Input schema for the rank_genes_groups plotting tool."""
     
     n_genes: int = Field(
@@ -631,7 +631,7 @@ class RankGenesGroupsModel(BaseVisualizationModel):
 
 
 # 重构 ClusterMapModel
-class ClusterMapModel(BaseModel):
+class ClusterMapParams(BaseModel):
     """Input schema for the clustermap plotting tool."""
 
     obs_keys: Optional[str] = Field(
@@ -645,8 +645,8 @@ class ClusterMapModel(BaseModel):
 
 
 
-# 重构 StackedViolinModel
-class StackedViolinModel(BaseStatPlotModel):
+# 重构 StackedViolinParams
+class StackedViolinParams(BaseStatPlotParams):
     """Input schema for the stacked_violin plotting tool."""
     
     stripplot: bool = Field(
@@ -688,8 +688,8 @@ class StackedViolinModel(BaseStatPlotModel):
         return v
 
 
-# 重构 TrackingModel
-class TrackingModel(BaseVisualizationModel):
+# 重构 TrackingParams
+class TrackingParams(BaseVisualizationParams):
     """Input schema for the tracking plotting tool."""
     
     groupby: str = Field(
@@ -717,8 +717,8 @@ class TrackingModel(BaseVisualizationModel):
         return v
 
 
-# 重构 EmbeddingDensityModel
-class EmbeddingDensityModel(BaseEmbeddingModel):
+# 重构 EmbeddingDensityParams
+class EmbeddingDensityParams(BaseEmbeddingParams):
     """Input schema for the embedding_density plotting tool."""
     
     basis: str = Field(
@@ -751,7 +751,7 @@ class EmbeddingDensityModel(BaseEmbeddingModel):
         return v
 
 
-class PCAModel(BaseEmbeddingModel):
+class PCAParams(BaseEmbeddingParams):
     """Input schema for the PCA plotting tool."""
     
     annotate_var_explained: bool = Field(
@@ -761,22 +761,22 @@ class PCAModel(BaseEmbeddingModel):
 
 
 # 重构 UMAP 模型
-class UMAPModel(BaseEmbeddingModel):
+class UMAPParams(BaseEmbeddingParams):
     """Input schema for the UMAP plotting tool."""
-    # 所有需要的字段已经在 BaseEmbeddingModel 中定义
+    # 所有需要的字段已经在 BaseEmbeddingParams 中定义
 
 
 # 重构 TSNE 模型
-class TSNEModel(BaseEmbeddingModel):
+class TSNEParams(BaseEmbeddingParams):
     """Input schema for the TSNE plotting tool."""
-    # 所有需要的字段已经在 BaseEmbeddingModel 中定义
+    # 所有需要的字段已经在 BaseEmbeddingParams 中定义
 
-# 重构 DiffusionMapModel
-class DiffusionMapModel(BaseEmbeddingModel):
+# 重构 DiffusionMapParams
+class DiffusionMapParams(BaseEmbeddingParams):
     """Input schema for the diffusion map plotting tool."""
-    # 所有需要的字段已经在 BaseEmbeddingModel 中定义
+    # 所有需要的字段已经在 BaseEmbeddingParams 中定义
 
-class HighestExprGenesModel(BaseVisualizationModel):
+class HighestExprGenesParams(BaseVisualizationParams):
     """Input schema for the highest_expr_genes plotting tool."""
     
     n_top: int = Field(
@@ -803,7 +803,7 @@ class HighestExprGenesModel(BaseVisualizationModel):
         return v
 
 
-class HighlyVariableGenesModel(BaseVisualizationModel):
+class HighlyVariableGenesParams(BaseVisualizationParams):
     """Input schema for the highly_variable_genes plotting tool."""
     
     log: bool = Field(
@@ -816,7 +816,7 @@ class HighlyVariableGenesModel(BaseVisualizationModel):
         description="Whether to plot highly variable genes or all genes."
     )
 
-class PCAVarianceRatioModel(BaseVisualizationModel):
+class PCAVarianceRatioParams(BaseVisualizationParams):
     """Input schema for the pca_variance_ratio plotting tool."""
     
     n_pcs: int = Field(
@@ -839,7 +839,7 @@ class PCAVarianceRatioModel(BaseVisualizationModel):
 
 # ... existing code ...
 
-class RankGenesGroupsDotplotModel(BaseMatrixModel):
+class RankGenesGroupsDotplotParams(BaseMatrixParams):
     """Input schema for the rank_genes_groups_dotplot plotting tool."""
     
     groups: Optional[Union[str, List[str]]] = Field(
@@ -874,7 +874,7 @@ class RankGenesGroupsDotplotModel(BaseMatrixModel):
 
 
 
-class EmbeddingModel(BaseEmbeddingModel):
+class EmbeddingParams(BaseEmbeddingParams):
     """Input schema for the embedding plotting tool."""
     
     basis: str = Field(

scmcp_shared/schema/pp.py

@@ -73,7 +73,7 @@ class FilterGenes(BaseModel):
         return v
 
 
-class SubsetCellModel(BaseModel):
+class SubsetCellParams(BaseModel):
     """Input schema for subsetting AnnData objects based on various criteria."""
     obs_key: Optional[str] = Field(
         default=None,
@@ -109,7 +109,7 @@ class SubsetCellModel(BaseModel):
     )
 
 
-class SubsetGeneModel(BaseModel):
+class SubsetGeneParams(BaseModel):
     """Input schema for subsetting AnnData objects based on various criteria."""
     min_counts: Optional[int] = Field(
         default=None,
@@ -196,7 +196,7 @@ class CalculateQCMetrics(BaseModel):
     
 
 
-class Log1PModel(BaseModel):
+class Log1PParams(BaseModel):
     """Input schema for the log1p preprocessing tool."""
     
     base: Optional[Union[int, float]] = Field(
@@ -233,7 +233,7 @@ class Log1PModel(BaseModel):
 
 
 
-class HighlyVariableGenesModel(BaseModel):
+class HighlyVariableGenesParams(BaseModel):
     """Input schema for the highly_variable_genes preprocessing tool."""
     
     layer: Optional[str] = Field(
@@ -307,7 +307,7 @@ class HighlyVariableGenesModel(BaseModel):
         return v
 
 
-class RegressOutModel(BaseModel):
+class RegressOutParams(BaseModel):
     """Input schema for the regress_out preprocessing tool."""
     
     keys: Union[str, List[str]] = Field(
@@ -340,7 +340,7 @@ class RegressOutModel(BaseModel):
         raise ValueError("keys must be a string or list of strings")
 
 
-class ScaleModel(BaseModel):
+class ScaleParams(BaseModel):
     """Input schema for the scale preprocessing tool."""
     
     zero_center: bool = Field(
@@ -376,7 +376,7 @@ class ScaleModel(BaseModel):
         return v
 
 
-class CombatModel(BaseModel):
+class CombatParams(BaseModel):
     """Input schema for the combat batch effect correction tool."""
     
     key: str = Field(
@@ -405,7 +405,7 @@ class CombatModel(BaseModel):
         return v
 
 
-class ScrubletModel(BaseModel):
+class ScrubletParams(BaseModel):
     """Input schema for the scrublet doublet prediction tool."""
     
     adata_sim: Optional[str] = Field(
@@ -511,7 +511,7 @@ class ScrubletModel(BaseModel):
         return v.lower()
 
 
-class NeighborsModel(BaseModel):
+class NeighborsParams(BaseModel):
     """Input schema for the neighbors graph construction tool."""
     
     n_neighbors: int = Field(
@@ -589,7 +589,7 @@ class NeighborsModel(BaseModel):
         return v
 
 
-class NormalizeTotalModel(BaseModel):
+class NormalizeTotalParams(BaseModel):
     """Input schema for the normalize_total preprocessing tool."""
     
     target_sum: Optional[float] = Field(

scmcp_shared/schema/tl.py

@@ -2,7 +2,7 @@ from pydantic import  Field, field_validator, ValidationInfo, BaseModel
 from typing import Optional, Union, List, Dict, Any, Tuple, Literal, Mapping
 
 
-class TSNEModel(BaseModel):
+class TSNEParams(BaseModel):
     """Input schema for the t-SNE dimensionality reduction tool."""
     n_pcs: Optional[int] = Field(
         default=None,
@@ -59,7 +59,7 @@ class TSNEModel(BaseModel):
         return v.lower()
 
 
-class UMAPModel(BaseModel):
+class UMAPParams(BaseModel):
     """Input schema for the UMAP dimensionality reduction tool."""
     
     min_dist: Optional[float] = Field(
@@ -145,7 +145,7 @@ class UMAPModel(BaseModel):
         return v.lower()
 
 
-class DrawGraphModel(BaseModel):
+class DrawGraphParams(BaseModel):
     """Input schema for the force-directed graph drawing tool."""
     
     layout: str = Field(
@@ -199,7 +199,7 @@ class DrawGraphModel(BaseModel):
         return v
 
 
-class DiffMapModel(BaseModel):
+class DiffMapParams(BaseModel):
     """Input schema for the Diffusion Maps dimensionality reduction tool."""
     
     n_comps: int = Field(
@@ -229,7 +229,7 @@ class DiffMapModel(BaseModel):
         return v
 
 
-class EmbeddingDensityModel(BaseModel):
+class EmbeddingDensityParams(BaseModel):
     """Input schema for the embedding density calculation tool."""
     
     basis: str = Field(
@@ -257,7 +257,7 @@ class EmbeddingDensityModel(BaseModel):
         return v
 
 
-class LeidenModel(BaseModel):
+class LeidenParams(BaseModel):
     """Input schema for the Leiden clustering algorithm."""
     
     resolution: Optional[float] = Field(
@@ -324,7 +324,7 @@ class LeidenModel(BaseModel):
         return v
 
 
-class LouvainModel(BaseModel):
+class LouvainParams(BaseModel):
     """Input schema for the Louvain clustering algorithm."""
     
     resolution: Optional[float] = Field(
@@ -396,7 +396,7 @@ class LouvainModel(BaseModel):
         return v
 
 
-class DendrogramModel(BaseModel):
+class DendrogramParams(BaseModel):
     """Input schema for the hierarchical clustering dendrogram tool."""
     
     groupby: str = Field(
@@ -461,7 +461,7 @@ class DendrogramModel(BaseModel):
         return v
 
 
-class DPTModel(BaseModel):
+class DPTParams(BaseModel):
     """Input schema for the Diffusion Pseudotime (DPT) tool."""
     
     n_dcs: int = Field(
@@ -510,7 +510,7 @@ class DPTModel(BaseModel):
             raise ValueError("min_group_size must be between 0 and 1")
         return v
 
-class PAGAModel(BaseModel):
+class PAGAParams(BaseModel):
     """Input schema for the Partition-based Graph Abstraction (PAGA) tool."""
     
     groups: Optional[str] = Field(
@@ -531,14 +531,14 @@ class PAGAModel(BaseModel):
     )
     
     @field_validator('model')
-    def validate_model(cls, v: str) -> str:
+    def validate_Params(cls, v: str) -> str:
         """Validate model version is supported"""
         if v not in ['v1.2', 'v1.0']:
             raise ValueError("model must be either 'v1.2' or 'v1.0'")
         return v
 
 
-class IngestModel(BaseModel):
+class IngestParams(BaseModel):
     """Input schema for the ingest tool that maps labels and embeddings from reference data to new data."""
     
     obs: Optional[Union[str, List[str]]] = Field(
@@ -587,7 +587,7 @@ class IngestModel(BaseModel):
         return v.lower()
 
 
-class RankGenesGroupsModel(BaseModel):
+class RankGenesGroupsParams(BaseModel):
     """Input schema for the rank_genes_groups tool."""
     
     groupby: str = Field(
@@ -669,7 +669,7 @@ class RankGenesGroupsModel(BaseModel):
         return v
 
 
-class FilterRankGenesGroupsModel(BaseModel):
+class FilterRankGenesGroupsParams(BaseModel):
     """Input schema for filtering ranked genes groups."""
     
     key: Optional[str] = Field(
@@ -732,7 +732,7 @@ class FilterRankGenesGroupsModel(BaseModel):
         return v
 
 
-class MarkerGeneOverlapModel(BaseModel):
+class MarkerGeneOverlapParams(BaseModel):
     """Input schema for the marker gene overlap tool."""
     
     key: str = Field(
@@ -803,7 +803,7 @@ class MarkerGeneOverlapModel(BaseModel):
         return v
 
 
-class ScoreGenesModel(BaseModel):
+class ScoreGenesParams(BaseModel):
     """Input schema for the score_genes tool that calculates gene scores based on average expression."""
     
     ctrl_size: int = Field(
@@ -846,7 +846,7 @@ class ScoreGenesModel(BaseModel):
         return v
 
 
-class ScoreGenesCellCycleModel(BaseModel):
+class ScoreGenesCellCycleParams(BaseModel):
     """Input schema for the score_genes_cell_cycle tool that scores cell cycle genes."""
     
     s_genes: List[str] = Field(
@@ -896,7 +896,7 @@ class ScoreGenesCellCycleModel(BaseModel):
 
 
 
-class PCAModel(BaseModel):
+class PCAParams(BaseModel):
     """Input schema for the PCA preprocessing tool."""
     
     n_comps: Optional[int] = Field(

scmcp_shared/schema/tool.py

@@ -0,0 +1,11 @@
+from pydantic import BaseModel, Field
+from typing import List
+
+
+class Tool(BaseModel):
+    name: str = Field(description="The name of the tool")
+    description: str = Field(description="The description of the tool")
+
+
+class ToolList(BaseModel):
+    tools: List[Tool]

scmcp_shared/schema/util.py

@@ -10,7 +10,7 @@ from typing import Optional, Union, List, Dict, Any, Callable, Collection, Liter
 
 
 
-class MarkVarModel(BaseModel):
+class MarkVarParams(BaseModel):
     """Determine or mark if each gene meets specific conditions and store results in adata.var as boolean values"""
     
     var_name: str = Field(
@@ -32,15 +32,15 @@ class MarkVarModel(BaseModel):
     )
 
 
-class ListVarModel(BaseModel):
+class ListVarParams(BaseModel):
     """ListVarModel"""    
     pass
 
-class ListObsModel(BaseModel):
+class ListObsParams(BaseModel):
     """ListObsModel"""    
     pass    
 
-class VarNamesModel(BaseModel):
+class VarNamesParams(BaseModel):
     """ListObsModel"""    
     var_names: List[str] = Field(
             default=None,
@@ -48,7 +48,7 @@ class VarNamesModel(BaseModel):
         )
 
 
-class ConcatBaseModel(BaseModel):
+class ConcatBaseParams(BaseModel):
     """Model for concatenating AnnData objects"""
     
     axis: Literal['obs', 0, 'var', 1] = Field(
@@ -89,7 +89,7 @@ class ConcatBaseModel(BaseModel):
     )
 
 
-class DPTIROOTModel(BaseModel):
+class DPTIROOTParams(BaseModel):
     """Input schema for setting the root cell for diffusion pseudotime."""
     diffmap_key: str = Field(
         default="X_diffmap",
@@ -103,7 +103,7 @@ class DPTIROOTModel(BaseModel):
     )
 
 
-class CelltypeMapCellTypeModel(BaseModel):
+class CelltypeMapCellTypeParams(BaseModel):
     """Input schema for mapping cluster IDs to cell type names."""
     cluster_key: str = Field(
         description="Key in adata.obs containing cluster IDs."
@@ -122,14 +122,14 @@ class CelltypeMapCellTypeModel(BaseModel):
     
 
 
-class AddLayerModel(BaseModel):
+class AddLayerParams(BaseModel):
     """Input schema for adding a layer to AnnData object."""
     layer_name: str = Field(
         description="Name of the layer to add to adata.layers."
     )
  
 
-class QueryOpLogModel(BaseModel):
+class QueryOpLogParams(BaseModel):
     """QueryOpLogModel"""    
     n: int = Field(
         default=10,