@tokenlabai/mcp-server 0.6.10 → 0.6.11

package/contract/openapi.json

@@ -21,6 +21,10 @@
       "name": "Chat",
       "description": "Chat completions API (OpenAI-compatible)"
     },
+    {
+      "name": "Responses",
+      "description": "OpenAI Responses API-compatible native endpoints"
+    },
     {
       "name": "Embeddings",
       "description": "Text embeddings API"
@@ -78,7 +82,7 @@
           "Chat"
         ],
         "summary": "Create chat completion",
-        "description": "Creates a model response for the given chat conversation. Supports streaming and function calling.",
+        "description": "Creates a model response from a Chat Completions request. Chat-native upstreams receive the request shape unchanged apart from routing and authentication fields; unknown fields are forwarded on a best-effort basis. Portable Chat requests may be compiled one-way to another protocol when a compatible physical endpoint is selected.",
         "operationId": "createChatCompletion",
         "requestBody": {
           "required": true,
@@ -2048,7 +2052,7 @@
           "Gemini"
         ],
         "summary": "Generate content (Gemini-compatible)",
-        "description": "Generates content using Google Gemini API format. Supports text and vision models.\n\n**Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
+        "description": "Generates content using the native Gemini GenerateContent shape. This route is exposed only when the model details advertise Gemini requests and a same-protocol route is currently available. ProtoJSON lowerCamelCase and original proto snake_case field names are preserved. Unknown fields are forwarded best-effort and support is determined by the selected provider.\n\n**Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
         "operationId": "geminiGenerateContent",
         "parameters": [
           {
@@ -2126,16 +2130,44 @@
             }
           },
           "400": {
-            "$ref": "#/components/responses/BadRequest"
+            "description": "Gemini-native invalid request error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "401": {
-            "$ref": "#/components/responses/Unauthorized"
+            "description": "Gemini-native authentication error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "402": {
-            "$ref": "#/components/responses/PaymentRequired"
+            "description": "Gemini-native payment error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "500": {
-            "$ref": "#/components/responses/ServerError"
+            "description": "Gemini-native server error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           }
         }
       }
@@ -2146,7 +2178,7 @@
           "Gemini"
         ],
         "summary": "Stream generate content (Gemini-compatible)",
-        "description": "Streams content generation using Google Gemini API format. Returns Server-Sent Events.\n\n**Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
+        "description": "Streams native Gemini GenerateContent Server-Sent Events without adding Chat Completions terminators. Metadata-only chunks, intermediate chunks without finishReason, and natural EOF are valid.\n\n**Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
         "operationId": "geminiStreamGenerateContent",
         "parameters": [
           {
@@ -2191,16 +2223,283 @@
             }
           },
           "400": {
-            "$ref": "#/components/responses/BadRequest"
+            "description": "Gemini-native invalid request error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "401": {
-            "$ref": "#/components/responses/Unauthorized"
+            "description": "Gemini-native authentication error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "402": {
-            "$ref": "#/components/responses/PaymentRequired"
+            "description": "Gemini-native payment error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           },
           "500": {
-            "$ref": "#/components/responses/ServerError"
+            "description": "Gemini-native server error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
+          }
+        }
+      }
+    },
+    "/v1beta/models/{model}:countTokens": {
+      "post": {
+        "tags": [
+          "Gemini"
+        ],
+        "summary": "Count tokens (Gemini-compatible)",
+        "description": "Counts tokens using the native Gemini CountTokens contract. Unknown fields and either official ProtoJSON field-name spelling are forwarded best-effort.",
+        "operationId": "geminiCountTokens",
+        "parameters": [
+          {
+            "name": "model",
+            "in": "path",
+            "required": true,
+            "schema": {
+              "type": "string"
+            }
+          },
+          {
+            "name": "key",
+            "in": "query",
+            "required": false,
+            "schema": {
+              "type": "string"
+            }
+          }
+        ],
+        "requestBody": {
+          "required": true,
+          "content": {
+            "application/json": {
+              "schema": {
+                "type": "object",
+                "properties": {
+                  "contents": {
+                    "type": "array",
+                    "items": {
+                      "type": "object",
+                      "additionalProperties": true
+                    }
+                  },
+                  "generateContentRequest": {
+                    "type": "object",
+                    "additionalProperties": true
+                  },
+                  "generate_content_request": {
+                    "type": "object",
+                    "additionalProperties": true
+                  }
+                },
+                "additionalProperties": true
+              }
+            }
+          }
+        },
+        "responses": {
+          "200": {
+            "description": "Gemini-native token count response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "type": "object",
+                  "additionalProperties": true
+                }
+              }
+            }
+          },
+          "400": {
+            "description": "Gemini-native error response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
+          }
+        }
+      }
+    },
+    "/v1beta/models/{model}:embedContent": {
+      "post": {
+        "tags": [
+          "Gemini"
+        ],
+        "summary": "Embed content (Gemini-compatible)",
+        "description": "Creates an embedding using the native Gemini EmbedContent contract.",
+        "operationId": "geminiEmbedContent",
+        "parameters": [
+          {
+            "name": "model",
+            "in": "path",
+            "required": true,
+            "schema": {
+              "type": "string"
+            }
+          },
+          {
+            "name": "key",
+            "in": "query",
+            "required": false,
+            "schema": {
+              "type": "string"
+            }
+          }
+        ],
+        "requestBody": {
+          "required": true,
+          "content": {
+            "application/json": {
+              "schema": {
+                "type": "object",
+                "required": [
+                  "content"
+                ],
+                "properties": {
+                  "content": {
+                    "type": "object",
+                    "additionalProperties": true
+                  },
+                  "taskType": {
+                    "type": "string"
+                  },
+                  "task_type": {
+                    "type": "string"
+                  },
+                  "title": {
+                    "type": "string"
+                  },
+                  "outputDimensionality": {
+                    "type": "integer"
+                  },
+                  "output_dimensionality": {
+                    "type": "integer"
+                  }
+                },
+                "additionalProperties": true
+              }
+            }
+          }
+        },
+        "responses": {
+          "200": {
+            "description": "Gemini-native embedding response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "type": "object",
+                  "additionalProperties": true
+                }
+              }
+            }
+          },
+          "400": {
+            "description": "Gemini-native error response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
+          }
+        }
+      }
+    },
+    "/v1beta/models/{model}:batchEmbedContents": {
+      "post": {
+        "tags": [
+          "Gemini"
+        ],
+        "summary": "Batch embed contents (Gemini-compatible)",
+        "description": "Creates multiple embeddings using the native Gemini BatchEmbedContents contract.",
+        "operationId": "geminiBatchEmbedContents",
+        "parameters": [
+          {
+            "name": "model",
+            "in": "path",
+            "required": true,
+            "schema": {
+              "type": "string"
+            }
+          },
+          {
+            "name": "key",
+            "in": "query",
+            "required": false,
+            "schema": {
+              "type": "string"
+            }
+          }
+        ],
+        "requestBody": {
+          "required": true,
+          "content": {
+            "application/json": {
+              "schema": {
+                "type": "object",
+                "required": [
+                  "requests"
+                ],
+                "properties": {
+                  "requests": {
+                    "type": "array",
+                    "items": {
+                      "type": "object",
+                      "additionalProperties": true
+                    }
+                  }
+                },
+                "additionalProperties": true
+              }
+            }
+          }
+        },
+        "responses": {
+          "200": {
+            "description": "Gemini-native batch embedding response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "type": "object",
+                  "additionalProperties": true
+                }
+              }
+            }
+          },
+          "400": {
+            "description": "Gemini-native error response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/GeminiErrorResponse"
+                }
+              }
+            }
           }
         }
       }
@@ -2208,10 +2507,10 @@
     "/v1/responses": {
       "post": {
         "tags": [
-          "Chat"
+          "Responses"
         ],
         "summary": "Create response (OpenAI Responses API)",
-        "description": "Creates a response using the OpenAI Responses API format. Supports stateful conversation management.",
+        "description": "Creates a response using the native OpenAI Responses API shape. The model details must advertise the Responses request format and a same-protocol route must be currently available; model names and providers do not imply availability. Unknown request fields are forwarded on a best-effort basis and remain subject to the selected provider's support.",
         "operationId": "createResponse",
         "requestBody": {
           "required": true,
@@ -2220,8 +2519,7 @@
               "schema": {
                 "type": "object",
                 "required": [
-                  "model",
-                  "input"
+                  "model"
                 ],
                 "properties": {
                   "model": {
@@ -2261,7 +2559,7 @@
                       "type": "object",
                       "additionalProperties": true
                     },
-                    "description": "Tools available to the model. For hosted image_generation tools that use the default image tool model or explicitly set model: gpt-image-2, TokenLab removes unsupported input_fidelity before forwarding because GPT Image 2 already treats image inputs as high fidelity. Do not send background: transparent for that tool; TokenLab does not silently remove it because that changes output semantics."
+                    "description": "Tools available to the model. Tool types and combinations are validated by the selected service."
                   },
                   "tool_choice": {
                     "oneOf": [
@@ -2283,13 +2581,9 @@
                         "description": "When true, request obfuscated tokens in Responses stream events from compatible upstreams."
                       }
                     },
-                    "additionalProperties": false,
+                    "additionalProperties": true,
                     "description": "Responses streaming options."
                   },
-                  "reasoning_effort": {
-                    "type": "string",
-                    "description": "Reasoning effort hint for compatible models."
-                  },
                   "include": {
                     "type": "array",
                     "items": {
@@ -2304,17 +2598,35 @@
                     ],
                     "description": "Service tier hint for compatible providers."
                   },
-                  "truncation_strategy": {
+                  "truncation": {
                     "type": "string",
-                    "description": "Truncation strategy for long conversations when supported."
+                    "description": "Truncation strategy for long conversations."
                   },
-                  "seed": {
-                    "type": "integer",
-                    "description": "Seed for deterministic-compatible providers."
+                  "background": {
+                    "type": "boolean",
+                    "description": "Whether to run the response asynchronously."
                   },
-                  "user": {
+                  "previous_response_id": {
                     "type": "string",
-                    "description": "End-user identifier."
+                    "description": "ID of a previous response to continue."
+                  },
+                  "prompt": {
+                    "type": "object",
+                    "additionalProperties": true,
+                    "description": "Reference to a reusable prompt template and variables."
+                  },
+                  "store": {
+                    "type": "boolean",
+                    "description": "Whether the response is stored for later retrieval."
+                  },
+                  "reasoning": {
+                    "type": "object",
+                    "additionalProperties": true,
+                    "description": "Reasoning configuration."
+                  },
+                  "top_p": {
+                    "type": "number",
+                    "description": "Nucleus sampling probability."
                   },
                   "parallel_tool_calls": {
                     "type": "boolean",
@@ -2330,7 +2642,8 @@
                     "additionalProperties": true,
                     "description": "Text formatting options."
                   }
-                }
+                },
+                "additionalProperties": true
               }
             }
           }
@@ -2351,7 +2664,7 @@
     "/v1/responses/{id}": {
       "get": {
         "tags": [
-          "Chat"
+          "Responses"
         ],
         "summary": "Retrieve response",
         "description": "Retrieves a stored Responses API response by TokenLab response id. Ownership checks return 404 when the response is missing or unavailable to the current API key.",
@@ -2365,6 +2678,41 @@
               "type": "string"
             },
             "description": "Response id returned by POST /v1/responses."
+          },
+          {
+            "name": "include",
+            "in": "query",
+            "schema": {
+              "type": "array",
+              "items": {
+                "type": "string"
+              }
+            },
+            "description": "Additional response fields to include."
+          },
+          {
+            "name": "include_obfuscation",
+            "in": "query",
+            "schema": {
+              "type": "boolean"
+            },
+            "description": "Whether streamed events include obfuscation fields."
+          },
+          {
+            "name": "stream",
+            "in": "query",
+            "schema": {
+              "type": "boolean"
+            },
+            "description": "Whether to retrieve the response as an event stream."
+          },
+          {
+            "name": "starting_after",
+            "in": "query",
+            "schema": {
+              "type": "integer"
+            },
+            "description": "Sequence number after which streaming events are returned."
           }
         ],
         "responses": {
@@ -2384,10 +2732,10 @@
       },
       "delete": {
         "tags": [
-          "Chat"
+          "Responses"
         ],
         "summary": "Delete response",
-        "description": "Deletes a stored Responses API response by TokenLab response id. When applicable, TokenLab clears the provider-side record before clearing local state.",
+        "description": "Deletes a stored Responses API response. This operation is deletion, not cancellation of an in-progress response.",
         "operationId": "deleteResponse",
         "parameters": [
           {
@@ -2419,10 +2767,10 @@
     "/v1/responses/compact": {
       "post": {
         "tags": [
-          "Chat"
+          "Responses"
         ],
         "summary": "Compact response context (OpenAI Responses API)",
-        "description": "Compacts a Responses conversation for Codex CLI remote compaction. This endpoint uses `POST /v1/responses/compact`; do not use `/v1/compact`. The request `model` is the current Codex session model and must be available on the Responses path.",
+        "description": "Compacts Responses input and returns the native `response.compaction` contract. The request model must be available on a same-protocol Responses path.",
         "operationId": "compactResponse",
         "requestBody": {
           "required": true,
@@ -6622,6 +6970,26 @@
               "$ref": "#/components/schemas/Tool"
             }
           },
+          "functions": {
+            "type": "array",
+            "description": "Deprecated function definitions retained by the official Chat Completions contract.",
+            "items": {
+              "type": "object",
+              "additionalProperties": true
+            }
+          },
+          "function_call": {
+            "oneOf": [
+              {
+                "type": "string"
+              },
+              {
+                "type": "object",
+                "additionalProperties": true
+              }
+            ],
+            "description": "Deprecated function-call selection retained by the official Chat Completions contract."
+          },
           "tool_choice": {
             "oneOf": [
               {
@@ -6705,11 +7073,6 @@
             "maximum": 20,
             "description": "Number of most likely tokens to return at each position when logprobs is enabled."
           },
-          "top_k": {
-            "type": "integer",
-            "minimum": 0,
-            "description": "Top-k sampling cutoff for compatible providers."
-          },
           "logit_bias": {
             "type": "object",
             "additionalProperties": {
@@ -6746,18 +7109,6 @@
             },
             "description": "Language settings for models that translate audio or video through Chat Completions."
           },
-          "speech_rate": {
-            "type": "number",
-            "minimum": 0.5,
-            "maximum": 2,
-            "description": "Speech rate for compatible translated audio output."
-          },
-          "repetition_penalty": {
-            "type": "number",
-            "minimum": 0,
-            "exclusiveMinimum": true,
-            "description": "Repetition penalty for compatible models."
-          },
           "prediction": {
             "type": "object",
             "additionalProperties": true,
@@ -6770,7 +7121,8 @@
             ],
             "description": "Service tier hint for compatible providers."
           }
-        }
+        },
+        "additionalProperties": true
       },
       "ChatMessage": {
         "type": "object",
@@ -9169,6 +9521,7 @@
       },
       "GeminiRequest": {
         "type": "object",
+        "description": "Native Gemini GenerateContent request. ProtoJSON lowerCamelCase and original proto snake_case names are both accepted and preserved, including mixed requests. If both spellings of one field are present, TokenLab does not merge them or choose a local precedence. Unknown fields are forwarded on a best-effort basis.",
         "required": [
           "contents"
         ],
@@ -9417,7 +9770,7 @@
               "candidateCount": {
                 "type": "integer",
                 "minimum": 1,
-                "description": "Number of response candidates for non-streaming generation. Streaming requests must omit this field or keep it at 1."
+                "description": "Requested number of response candidates. Support is determined by the selected service."
               },
               "responseModalities": {
                 "type": "array",
@@ -9473,7 +9826,7 @@
           },
           "tools": {
             "type": "array",
-            "description": "Gemini tools. Supported tool combinations depend on the selected model and route; unsupported native image-output tool combinations are rejected before upstream retries.",
+            "description": "Gemini tools. Tool types and combinations are interpreted by the selected service.",
             "items": {
               "type": "object",
               "properties": {
@@ -9575,8 +9928,36 @@
           "cachedContent": {
             "type": "string",
             "description": "Gemini cached content resource name for compatible models."
+          },
+          "system_instruction": {
+            "type": "object",
+            "additionalProperties": true,
+            "description": "Original proto field-name spelling of systemInstruction."
+          },
+          "generation_config": {
+            "type": "object",
+            "additionalProperties": true,
+            "description": "Original proto field-name spelling of generationConfig."
+          },
+          "safety_settings": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "additionalProperties": true
+            },
+            "description": "Original proto field-name spelling of safetySettings."
+          },
+          "tool_config": {
+            "type": "object",
+            "additionalProperties": true,
+            "description": "Original proto field-name spelling of toolConfig."
+          },
+          "cached_content": {
+            "type": "string",
+            "description": "Original proto field-name spelling of cachedContent."
           }
-        }
+        },
+        "additionalProperties": true
       },
       "GeminiResponse": {
         "type": "object",
@@ -10852,6 +11233,52 @@
       },
       "GeminiModelInfo": {
         "type": "object",
+        "required": [
+          "name",
+          "baseModelId",
+          "version"
+        ],
+        "properties": {
+          "name": {
+            "type": "string"
+          },
+          "baseModelId": {
+            "type": "string"
+          },
+          "version": {
+            "type": "string"
+          },
+          "displayName": {
+            "type": "string"
+          },
+          "description": {
+            "type": "string"
+          },
+          "inputTokenLimit": {
+            "type": "integer"
+          },
+          "outputTokenLimit": {
+            "type": "integer"
+          },
+          "supportedGenerationMethods": {
+            "type": "array",
+            "items": {
+              "type": "string"
+            }
+          },
+          "temperature": {
+            "type": "number"
+          },
+          "maxTemperature": {
+            "type": "number"
+          },
+          "topP": {
+            "type": "number"
+          },
+          "topK": {
+            "type": "number"
+          }
+        },
         "additionalProperties": true
       },
       "GeminiModelListResponse": {