mcp-embed-cohere mcp

v1.0.0 · MCP Tool · ai · registry.pascalai.org

Generate text embeddings using Cohere Embed API for semantic search and similarity tasks.

cohereembeddingsvectorssimilaritynlp

Input Parameters

Parameter	Type	Description
api_keyrequired	string
modeloptional	string	Default: `embed-english-v3.0`.
textsrequired	array[string]
input_typeoptional	string	One of: `search_document`, `search_query`, `classification`, `clustering`. Default: `search_document`.
truncateoptional	string	One of: `NONE`, `START`, `END`. Default: `END`.
embedding_typesoptional	array[string]	Default: `['float']`.

Output Fields

Field	Type	Description
embeddings	array[array]
model	string
token_count	integer

Examples

Embed documents for search

// Input
{
  "api_key": "xxx",
  "model": "embed-english-v3.0",
  "texts": [
    "Paris is the capital of France",
    "London is in England"
  ],
  "input_type": "search_document"
}

// Output
{
  "embeddings": [
    [
      0.1,
      0.2
    ],
    [
      0.3,
      0.4
    ]
  ],
  "model": "embed-english-v3.0"
}

Install & Discovery

Install

ppm install mcp-embed-cohere

Get JSON Schema

GET /v1/packages/mcp-embed-cohere/1.0.0/schema

Discover by keyword

GET /v1/mcp/discover?q=cohere

Discovery hint: Install with ppm install mcp-embed-cohere or invoke remotely via POST /v1/invoke/mcp-embed-cohere on the MCP Service.

PascalAI Usage

uses toolslib;
var Tool := LoadTool('mcp-embed-cohere');
var R := Tool.Call(JsonObj([]));
Writeln(R.ToJSON);