@tokenlabai/mcp-server 0.6.10 → 0.6.12

package/generated/public-contract.json

@@ -5,7 +5,7 @@
     "id": "tokenlab-mcp-server",
     "name": "@tokenlabai/mcp-server",
     "title": "TokenLab MCP Server",
-    "version": "0.6.10",
+    "version": "0.6.12",
     "registry_name": "io.github.hedging8563/tokenlab",
     "source_url": "https://github.com/hedging8563/tokenlab-mcp-server",
     "landing_url": "https://tokenlab.sh/mcp",
@@ -25,8 +25,8 @@
     "openapi": "3.0.3",
     "title": "TokenLab AI Gateway",
     "version": "1.0.0",
-    "sha256": "40661266d5b182d7a9f829ae480a1c86d5f141c2b75b25f40ba4bfdf50a42c08",
-    "tool_manifest_sha256": "7d609ec553e0ff57721fd45637d720afda1079d8c99ec17c29b13b7d02b39d50"
+    "sha256": "a5d2c856f4c346924787a0870cf16ad748f54a51faeed9627d78d88ba20501ab",
+    "tool_manifest_sha256": "b64c0aea3103963b9c686344bcec839478ffb3b1e34b14ed91bcf5684826ee52"
   },
   "profiles": {
     "catalog": {
@@ -84,9 +84,9 @@
     },
     "full": {
       "is_default": false,
-      "endpoint_tools": 75,
+      "endpoint_tools": 78,
       "composite_tools": 2,
-      "total_tools": 77,
+      "total_tools": 80,
       "tool_names": [
         "cancel_anthropic_message_batch",
         "cancel_batch",
@@ -125,6 +125,9 @@
         "dispatch_volc_compatible_seedance_action_post",
         "edit_image",
         "edit_image_file",
+        "gemini_batch_embed_contents",
+        "gemini_count_tokens",
+        "gemini_embed_content",
         "get_3_dtask",
         "get_api_overview",
         "get_gemini_model",

package/generated/tools.json

@@ -6,7 +6,7 @@
     "openapi": "3.0.3",
     "title": "TokenLab AI Gateway",
     "version": "1.0.0",
-    "sha256": "40661266d5b182d7a9f829ae480a1c86d5f141c2b75b25f40ba4bfdf50a42c08"
+    "sha256": "a5d2c856f4c346924787a0870cf16ad748f54a51faeed9627d78d88ba20501ab"
   },
   "default_profile": "core",
   "profiles": [
@@ -14,7 +14,7 @@
     "core",
     "full"
   ],
-  "tool_count": 75,
+  "tool_count": 78,
   "tools": [
     {
       "name": "cancel_anthropic_message_batch",
@@ -174,12 +174,12 @@
       "content_type": "application/json",
       "auth": "required",
       "tags": [
-        "Chat"
+        "Responses"
       ],
       "profiles": [
         "full"
       ],
-      "description": "Compact response context (OpenAI Responses API) 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": "Compact response context (OpenAI Responses API) Compacts Responses input and returns the native `response.compaction` contract. The request model must be available on a same-protocol Responses path.",
       "input_schema": {
         "type": "object",
         "properties": {
@@ -1031,7 +1031,7 @@
         "core",
         "full"
       ],
-      "description": "Create chat completion Creates a model response for the given chat conversation. Supports streaming and function calling.",
+      "description": "Create an OpenAI-compatible chat completion. For inline image_url data URLs, declare the byte-accurate image MIME type. The MCP boundary corrects recognized PNG, JPEG, WebP, and GIF payloads declared as application/octet-stream and rejects unrecognized generic binary image payloads before sending a billable request.",
       "input_schema": {
         "type": "object",
         "properties": {
@@ -1344,6 +1344,26 @@
               }
             }
           },
+          "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": [
               {
@@ -1427,11 +1447,6 @@
             "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": {
@@ -1468,18 +1483,6 @@
             },
             "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,
@@ -1516,6 +1519,8 @@
           "presence_penalty",
           "frequency_penalty",
           "tools",
+          "functions",
+          "function_call",
           "tool_choice",
           "response_format",
           "seed",
@@ -1525,13 +1530,10 @@
           "reasoning_effort",
           "logprobs",
           "top_logprobs",
-          "top_k",
           "logit_bias",
           "modalities",
           "audio",
           "translation_options",
-          "speech_rate",
-          "repetition_penalty",
           "prediction",
           "service_tier"
         ],
@@ -1656,7 +1658,7 @@
         "core",
         "full"
       ],
-      "description": "Generate content (Gemini-compatible) Generates content using Google Gemini API format. Supports text and vision models. **Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
+      "description": "Generate content (Gemini-compatible) 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. **Authentication:** Use `?key=` query parameter, `x-goog-api-key` header, or `Authorization: Bearer` header.",
       "input_schema": {
         "type": "object",
         "properties": {
@@ -1908,7 +1910,7 @@
               "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",
@@ -1964,7 +1966,7 @@
           },
           "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": {
@@ -2066,6 +2068,33 @@
           "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."
           }
         },
         "required": [
@@ -2087,7 +2116,12 @@
           "safetySettings",
           "tools",
           "toolConfig",
-          "cachedContent"
+          "cachedContent",
+          "system_instruction",
+          "generation_config",
+          "safety_settings",
+          "tool_config",
+          "cached_content"
         ],
         "files": []
       },
@@ -2850,13 +2884,13 @@
       "content_type": "application/json",
       "auth": "required",
       "tags": [
-        "Chat"
+        "Responses"
       ],
       "profiles": [
         "core",
         "full"
       ],
-      "description": "Create response (OpenAI Responses API) Creates a response using the OpenAI Responses API format. Supports stateful conversation management.",
+      "description": "Create response (OpenAI Responses API) 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.",
       "input_schema": {
         "type": "object",
         "properties": {
@@ -2899,7 +2933,7 @@
               "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": [
@@ -2921,13 +2955,9 @@
                 "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": {
@@ -2942,17 +2972,35 @@
             ],
             "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",
@@ -2970,8 +3018,7 @@
           }
         },
         "required": [
-          "model",
-          "input"
+          "model"
         ],
         "additionalProperties": false
       },
@@ -2989,12 +3036,15 @@
           "tools",
           "tool_choice",
           "stream_options",
-          "reasoning_effort",
           "include",
           "service_tier",
-          "truncation_strategy",
-          "seed",
-          "user",
+          "truncation",
+          "background",
+          "previous_response_id",
+          "prompt",
+          "store",
+          "reasoning",
+          "top_p",
           "parallel_tool_calls",
           "metadata",
           "text"
@@ -4358,12 +4408,12 @@
       "content_type": null,
       "auth": "required",
       "tags": [
-        "Chat"
+        "Responses"
       ],
       "profiles": [
         "full"
       ],
-      "description": "Delete response Deletes a stored Responses API response by TokenLab response id. When applicable, TokenLab clears the provider-side record before clearing local state.",
+      "description": "Delete response Deletes a stored Responses API response. This operation is deletion, not cancellation of an in-progress response.",
       "input_schema": {
         "type": "object",
         "properties": {
@@ -5266,6 +5316,206 @@
         ]
       }
     },
+    {
+      "name": "gemini_batch_embed_contents",
+      "title": "Batch embed contents (Gemini-compatible)",
+      "operation_id": "geminiBatchEmbedContents",
+      "method": "POST",
+      "path": "/v1beta/models/{model}:batchEmbedContents",
+      "content_type": "application/json",
+      "auth": "required",
+      "tags": [
+        "Gemini"
+      ],
+      "profiles": [
+        "full"
+      ],
+      "description": "Batch embed contents (Gemini-compatible) Creates multiple embeddings using the native Gemini BatchEmbedContents contract.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "model": {
+            "type": "string"
+          },
+          "key": {
+            "type": "string"
+          },
+          "requests": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "additionalProperties": true
+            }
+          }
+        },
+        "required": [
+          "model",
+          "requests"
+        ],
+        "additionalProperties": false
+      },
+      "bindings": {
+        "path": [
+          "model"
+        ],
+        "query": [
+          "key"
+        ],
+        "header": [],
+        "body": [
+          "requests"
+        ],
+        "files": []
+      },
+      "annotations": {
+        "readOnlyHint": false,
+        "destructiveHint": false,
+        "idempotentHint": false,
+        "openWorldHint": true
+      }
+    },
+    {
+      "name": "gemini_count_tokens",
+      "title": "Count tokens (Gemini-compatible)",
+      "operation_id": "geminiCountTokens",
+      "method": "POST",
+      "path": "/v1beta/models/{model}:countTokens",
+      "content_type": "application/json",
+      "auth": "required",
+      "tags": [
+        "Gemini"
+      ],
+      "profiles": [
+        "full"
+      ],
+      "description": "Count tokens (Gemini-compatible) Counts tokens using the native Gemini CountTokens contract. Unknown fields and either official ProtoJSON field-name spelling are forwarded best-effort.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "model": {
+            "type": "string"
+          },
+          "key": {
+            "type": "string"
+          },
+          "contents": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "additionalProperties": true
+            }
+          },
+          "generateContentRequest": {
+            "type": "object",
+            "additionalProperties": true
+          },
+          "generate_content_request": {
+            "type": "object",
+            "additionalProperties": true
+          }
+        },
+        "required": [
+          "model"
+        ],
+        "additionalProperties": false
+      },
+      "bindings": {
+        "path": [
+          "model"
+        ],
+        "query": [
+          "key"
+        ],
+        "header": [],
+        "body": [
+          "contents",
+          "generateContentRequest",
+          "generate_content_request"
+        ],
+        "files": []
+      },
+      "annotations": {
+        "readOnlyHint": false,
+        "destructiveHint": false,
+        "idempotentHint": false,
+        "openWorldHint": true
+      }
+    },
+    {
+      "name": "gemini_embed_content",
+      "title": "Embed content (Gemini-compatible)",
+      "operation_id": "geminiEmbedContent",
+      "method": "POST",
+      "path": "/v1beta/models/{model}:embedContent",
+      "content_type": "application/json",
+      "auth": "required",
+      "tags": [
+        "Gemini"
+      ],
+      "profiles": [
+        "full"
+      ],
+      "description": "Embed content (Gemini-compatible) Creates an embedding using the native Gemini EmbedContent contract.",
+      "input_schema": {
+        "type": "object",
+        "properties": {
+          "model": {
+            "type": "string"
+          },
+          "key": {
+            "type": "string"
+          },
+          "content": {
+            "type": "object",
+            "additionalProperties": true
+          },
+          "taskType": {
+            "type": "string"
+          },
+          "task_type": {
+            "type": "string"
+          },
+          "title": {
+            "type": "string"
+          },
+          "outputDimensionality": {
+            "type": "integer"
+          },
+          "output_dimensionality": {
+            "type": "integer"
+          }
+        },
+        "required": [
+          "model",
+          "content"
+        ],
+        "additionalProperties": false
+      },
+      "bindings": {
+        "path": [
+          "model"
+        ],
+        "query": [
+          "key"
+        ],
+        "header": [],
+        "body": [
+          "content",
+          "taskType",
+          "task_type",
+          "title",
+          "outputDimensionality",
+          "output_dimensionality"
+        ],
+        "files": []
+      },
+      "annotations": {
+        "readOnlyHint": false,
+        "destructiveHint": false,
+        "idempotentHint": false,
+        "openWorldHint": true
+      }
+    },
     {
       "name": "get_3_dtask",
       "title": "Get 3D task status",
@@ -7130,7 +7380,7 @@
       "content_type": null,
       "auth": "required",
       "tags": [
-        "Chat"
+        "Responses"
       ],
       "profiles": [
         "full"
@@ -7142,6 +7392,25 @@
           "id": {
             "type": "string",
             "description": "Response id returned by POST /v1/responses."
+          },
+          "include": {
+            "type": "array",
+            "items": {
+              "type": "string"
+            },
+            "description": "Additional response fields to include."
+          },
+          "include_obfuscation": {
+            "type": "boolean",
+            "description": "Whether streamed events include obfuscation fields."
+          },
+          "stream": {
+            "type": "boolean",
+            "description": "Whether to retrieve the response as an event stream."
+          },
+          "starting_after": {
+            "type": "integer",
+            "description": "Sequence number after which streaming events are returned."
           }
         },
         "required": [
@@ -7153,7 +7422,12 @@
         "path": [
           "id"
         ],
-        "query": [],
+        "query": [
+          "include",
+          "include_obfuscation",
+          "stream",
+          "starting_after"
+        ],
         "header": [],
         "body": [],
         "files": []

package/llms-install.md

@@ -34,7 +34,7 @@ Public model catalog and pricing tools work without credentials. When inference
 }
 ```
 
-Set `TOKENLAB_MCP_TOOL_PROFILE=catalog` when the client should expose only the six public discovery tools. The default `core` profile exposes 31 tools, and `full` exposes 78.
+Set `TOKENLAB_MCP_TOOL_PROFILE=catalog` when the client should expose only the six public discovery tools. The default `core` profile exposes 31 tools, and `full` exposes 80.
 
 ## Verification
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@tokenlabai/mcp-server",
-  "version": "0.6.10",
+  "version": "0.6.12",
   "description": "OpenAPI-generated TokenLab MCP server with catalog, native AI, multimodal, resource, prompt, and async task support.",
   "type": "module",
   "engines": {
@@ -18,15 +18,17 @@
     "llms-install.md",
     "scripts/generate-contract.mjs",
     "scripts/sync-contract.mjs",
-    "src/index.js"
+    "src/index.js",
+    "src/media-mime.js"
   ],
   "scripts": {
     "start": "node ./src/index.js",
     "contract:generate": "node ./scripts/generate-contract.mjs",
     "contract:check": "node ./scripts/generate-contract.mjs --check",
+    "contract:source-check": "node ./scripts/sync-contract.mjs --check",
     "contract:sync": "node ./scripts/sync-contract.mjs",
     "version": "node ./scripts/sync-version.mjs && npm run contract:generate",
-    "test": "node --check ./src/index.js && node --check ./scripts/generate-contract.mjs && npm run contract:check && node --test"
+    "test": "node --check ./src/index.js && node --check ./src/media-mime.js && node --check ./scripts/generate-contract.mjs && node --check ./scripts/sync-contract.mjs && npm run contract:check && node --test"
   },
   "keywords": [
     "mcp",