@oomol-lab/open-connector 1.6.3 → 1.6.5

package/assets/open-connector/catalog/apps/pdf_vector.json

@@ -0,0 +1,329 @@
+{
+  "service": "pdf_vector",
+  "displayName": "PDF Vector",
+  "description": "Parse, question, and extract structured data from documents available at public URLs.",
+  "categories": [
+    "Documents",
+    "Productivity",
+    "AI"
+  ],
+  "authTypes": [
+    "api_key"
+  ],
+  "auth": [
+    {
+      "type": "api_key",
+      "label": "API Key",
+      "placeholder": "PDFVECTOR_API_KEY",
+      "description": "PDF Vector API key sent as a Bearer token. Create an account and manage your API key in the dashboard at https://app.pdfvector.com.",
+      "extraFields": []
+    }
+  ],
+  "homepageUrl": "https://www.pdfvector.com",
+  "actions": [
+    {
+      "id": "pdf_vector.ask_document",
+      "service": "pdf_vector",
+      "name": "ask_document",
+      "description": "Ask a question about a document available at a public URL with PDF Vector.",
+      "operationType": "read",
+      "requiredScopes": [],
+      "providerPermissions": [],
+      "inputSchema": {
+        "type": "object",
+        "properties": {
+          "documentUrl": {
+            "type": "string",
+            "description": "The publicly accessible HTTP or HTTPS URL of the document that PDF Vector should process.",
+            "format": "uri"
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "auto",
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The PDF Vector model tier. Defaults to auto, which lets PDF Vector select a tier."
+          },
+          "documentId": {
+            "type": "string",
+            "minLength": 1,
+            "maxLength": 512,
+            "description": "An optional external document identifier recorded with PDF Vector usage."
+          },
+          "question": {
+            "type": "string",
+            "minLength": 4,
+            "description": "The question PDF Vector should answer from the document."
+          }
+        },
+        "additionalProperties": false,
+        "required": [
+          "documentUrl",
+          "question"
+        ],
+        "description": "A public document URL, question, and PDF Vector model options."
+      },
+      "outputSchema": {
+        "type": "object",
+        "properties": {
+          "markdown": {
+            "type": "string",
+            "description": "The Markdown answer to the question."
+          },
+          "pageCount": {
+            "type": "integer",
+            "description": "The number of pages processed by PDF Vector."
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The model tier used by PDF Vector."
+          },
+          "credits": {
+            "type": "integer",
+            "description": "The number of PDF Vector credits consumed by the request."
+          },
+          "requestId": {
+            "type": "integer",
+            "description": "The PDF Vector request identifier."
+          },
+          "documentId": {
+            "type": "string",
+            "description": "The external document identifier when one was supplied."
+          }
+        },
+        "additionalProperties": true,
+        "required": [
+          "markdown",
+          "pageCount",
+          "model",
+          "credits",
+          "requestId"
+        ],
+        "description": "The answer and usage metadata returned by PDF Vector."
+      }
+    },
+    {
+      "id": "pdf_vector.extract_document",
+      "service": "pdf_vector",
+      "name": "extract_document",
+      "description": "Extract structured JSON matching a supplied schema from a document at a public URL with PDF Vector.",
+      "operationType": "read",
+      "requiredScopes": [],
+      "providerPermissions": [],
+      "inputSchema": {
+        "type": "object",
+        "properties": {
+          "documentUrl": {
+            "type": "string",
+            "description": "The publicly accessible HTTP or HTTPS URL of the document that PDF Vector should process.",
+            "format": "uri"
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "auto",
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The PDF Vector model tier. Defaults to auto, which lets PDF Vector select a tier."
+          },
+          "documentId": {
+            "type": "string",
+            "minLength": 1,
+            "maxLength": 512,
+            "description": "An optional external document identifier recorded with PDF Vector usage."
+          },
+          "prompt": {
+            "type": "string",
+            "minLength": 4,
+            "description": "Instructions describing the data to extract from the document."
+          },
+          "schema": {
+            "type": "object",
+            "properties": {},
+            "additionalProperties": true,
+            "description": "The JSON Schema describing the structure PDF Vector should extract."
+          }
+        },
+        "additionalProperties": false,
+        "required": [
+          "documentUrl",
+          "prompt",
+          "schema"
+        ],
+        "description": "A public document URL, extraction prompt, JSON Schema, and PDF Vector model options."
+      },
+      "outputSchema": {
+        "type": "object",
+        "properties": {
+          "data": {
+            "description": "The structured data produced from the caller-supplied JSON Schema."
+          },
+          "pageCount": {
+            "type": "integer",
+            "description": "The number of pages processed by PDF Vector."
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The model tier used by PDF Vector."
+          },
+          "credits": {
+            "type": "integer",
+            "description": "The number of PDF Vector credits consumed by the request."
+          },
+          "requestId": {
+            "type": "integer",
+            "description": "The PDF Vector request identifier."
+          },
+          "documentId": {
+            "type": "string",
+            "description": "The external document identifier when one was supplied."
+          }
+        },
+        "additionalProperties": true,
+        "required": [
+          "pageCount",
+          "model",
+          "credits",
+          "requestId"
+        ],
+        "description": "The extracted JSON value and usage metadata returned by PDF Vector."
+      }
+    },
+    {
+      "id": "pdf_vector.parse_document",
+      "service": "pdf_vector",
+      "name": "parse_document",
+      "description": "Parse a document from a public URL into Markdown with PDF Vector.",
+      "operationType": "read",
+      "requiredScopes": [],
+      "providerPermissions": [],
+      "inputSchema": {
+        "type": "object",
+        "properties": {
+          "documentUrl": {
+            "type": "string",
+            "description": "The publicly accessible HTTP or HTTPS URL of the document that PDF Vector should process.",
+            "format": "uri"
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "auto",
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The PDF Vector model tier. Defaults to auto, which lets PDF Vector select a tier."
+          },
+          "documentId": {
+            "type": "string",
+            "minLength": 1,
+            "maxLength": 512,
+            "description": "An optional external document identifier recorded with PDF Vector usage."
+          },
+          "includePages": {
+            "type": "boolean",
+            "description": "Whether to include page-separated Markdown in addition to the full document text."
+          }
+        },
+        "additionalProperties": false,
+        "required": [
+          "documentUrl"
+        ],
+        "description": "A public document URL and PDF Vector parsing options."
+      },
+      "outputSchema": {
+        "type": "object",
+        "properties": {
+          "markdown": {
+            "type": "string",
+            "description": "The full Markdown content extracted from the document."
+          },
+          "pageCount": {
+            "type": "integer",
+            "description": "The number of pages processed by PDF Vector."
+          },
+          "model": {
+            "type": "string",
+            "enum": [
+              "nano",
+              "mini",
+              "pro",
+              "max"
+            ],
+            "description": "The model tier used by PDF Vector."
+          },
+          "credits": {
+            "type": "integer",
+            "description": "The number of PDF Vector credits consumed by the request."
+          },
+          "requestId": {
+            "type": "integer",
+            "description": "The PDF Vector request identifier."
+          },
+          "documentId": {
+            "type": "string",
+            "description": "The external document identifier when one was supplied."
+          },
+          "html": {
+            "type": "string",
+            "description": "The rich HTML representation returned when supported by the model."
+          },
+          "pages": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "properties": {
+                "pageNumber": {
+                  "type": "integer",
+                  "minimum": 1,
+                  "description": "The 1-based page number in the source document."
+                },
+                "markdown": {
+                  "type": "string",
+                  "description": "The Markdown content extracted from this page."
+                }
+              },
+              "additionalProperties": true,
+              "required": [
+                "pageNumber",
+                "markdown"
+              ],
+              "description": "The extracted content for one document page."
+            },
+            "description": "Page-separated Markdown when includePages is enabled."
+          }
+        },
+        "additionalProperties": true,
+        "required": [
+          "markdown",
+          "pageCount",
+          "model",
+          "credits",
+          "requestId"
+        ],
+        "description": "The document content and usage metadata returned by PDF Vector."
+      }
+    }
+  ]
+}