pinecone-mcp
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChinesePinecone MCP Tools Reference
Pinecone MCP 工具参考
The Pinecone MCP server exposes the following tools to AI agents and IDEs. For setup and installation instructions, see the MCP server guide.
Key Limitation: The Pinecone MCP only supports integrated indexes — indexes created with a built-in Pinecone embedding model. It does not work with standard indexes using external embedding models. For those, use the Pinecone CLI.
Pinecone MCP server 向 AI Agent 和 IDE 提供以下工具。有关设置和安装说明,请参阅 MCP server 指南。
关键限制: Pinecone MCP 仅支持集成索引——即使用内置 Pinecone embedding 模型创建的索引。它不支持使用外部 embedding 模型的标准索引。对于此类索引,请使用 Pinecone CLI。
list-indexes
list-indexeslist-indexes
list-indexesList all indexes in the current Pinecone project.
列出当前 Pinecone 项目中的所有索引。
describe-index
describe-indexdescribe-index
describe-indexGet configuration details for a specific index — cloud, region, dimension, metric, embedding model, field map, and status.
Parameters:
- (required) — Index name
name
获取指定索引的配置详情,包括云、区域、维度、度量指标、embedding 模型、字段映射和状态。
参数:
- (必填)— 索引名称
name
describe-index-stats
describe-index-statsdescribe-index-stats
describe-index-statsGet statistics for an index including total record count and per-namespace breakdown.
Parameters:
- (required) — Index name
name
获取索引的统计信息,包括总记录数以及按 namespace 细分的数据。
参数:
- (必填)— 索引名称
name
create-index-for-model
create-index-for-modelcreate-index-for-model
create-index-for-modelCreate a new serverless index with an integrated embedding model. Pinecone handles embedding automatically — no external model needed.
Parameters:
- (required) — Index name
name - (required) —
cloud,aws, orgcpazure - (required) — Cloud region (e.g.
region)us-east-1 - (required) — Embedding model:
embed.model,llama-text-embed-v2, ormultilingual-e5-largepinecone-sparse-english-v0 - (required) — The record field that contains text to embed (e.g.
embed.fieldMap.text)chunk_text
使用集成 embedding 模型创建新的 serverless 索引。Pinecone 会自动处理 embedding,无需使用外部模型。
参数:
- (必填)— 索引名称
name - (必填)—
cloud、aws或gcpazure - (必填)— 云区域(例如
region)us-east-1 - (必填)— Embedding 模型:
embed.model、llama-text-embed-v2或multilingual-e5-largepinecone-sparse-english-v0 - (必填)— 包含待 embedding 文本的记录字段(例如
embed.fieldMap.text)chunk_text
upsert-records
upsert-recordsupsert-records
upsert-recordsInsert or update records in an integrated index. Records are automatically embedded using the index's configured model.
Parameters:
- (required) — Index name
name - (required) — Namespace to upsert into
namespace - (required) — Array of records. Each record must have an
recordsoridfield and contain the text field specified in the index's_id. Do not nest fields underfieldMap— put them directly on the record.metadata
Example record:
json
{ "_id": "rec1", "chunk_text": "The Eiffel Tower was built in 1889.", "category": "architecture" }在集成索引中插入或更新记录。系统会使用索引配置的模型自动为记录生成 embedding。
参数:
- (必填)— 索引名称
name - (必填)— 要写入的 namespace
namespace - (必填)— 记录数组。每条记录必须包含
records或id字段,并包含索引_id中指定的文本字段。不要将字段嵌套在fieldMap下,应直接放在记录顶层。metadata
记录示例:
json
{ "_id": "rec1", "chunk_text": "The Eiffel Tower was built in 1889.", "category": "architecture" }search-records
search-recordssearch-records
search-recordsSemantic text search against an integrated index. Pass plain text — the MCP embeds the query automatically using the index's model.
Parameters:
- (required) — Index name
name - (required) — Namespace to search
namespace - (required) — The text query
query.inputs.text - (required) — Number of results to return
query.topK - (optional) — Metadata filter using MongoDB-style operators (
query.filter,$eq,$ne,$in,$gt,$gte,$lt)$lte - (optional) — Reranking model:
rerank.model,bge-reranker-v2-m3, orcohere-rerank-3.5pinecone-rerank-v0 - (optional) — Fields to rerank on (e.g.
rerank.rankFields)["chunk_text"] - (optional) — Number of results to return after reranking
rerank.topN
针对集成索引执行语义文本搜索。传入纯文本即可——MCP 会使用索引的模型自动为查询生成 embedding。
参数:
- (必填)— 索引名称
name - (必填)— 要搜索的 namespace
namespace - (必填)— 文本查询
query.inputs.text - (必填)— 返回的结果数量
query.topK - (可选)— 使用 MongoDB 风格运算符(
query.filter、$eq、$ne、$in、$gt、$gte、$lt)的元数据过滤器$lte - (可选)— 重排序模型:
rerank.model、bge-reranker-v2-m3或cohere-rerank-3.5pinecone-rerank-v0 - (可选)— 用于重排序的字段(例如
rerank.rankFields)["chunk_text"] - (可选)— 重排序后返回的结果数量
rerank.topN
cascading-search
cascading-searchcascading-search
cascading-searchSearch across multiple indexes simultaneously, then deduplicate and rerank results into a single ranked list.
Parameters:
- (required) — Array of
indexesobjects to search across{ name, namespace } - (required) — The text query
query.inputs.text - (required) — Number of results to retrieve per index before reranking
query.topK - (required) — Reranking model:
rerank.model,bge-reranker-v2-m3, orcohere-rerank-3.5pinecone-rerank-v0 - (required) — Fields to rerank on
rerank.rankFields - (optional) — Final number of results to return after reranking
rerank.topN
同时搜索多个索引,然后对结果去重并重排序,合并为单个排名列表。
参数:
- (必填)— 要搜索的
indexes对象数组{ name, namespace } - (必填)— 文本查询
query.inputs.text - (必填)— 重排序前从每个索引获取的结果数量
query.topK - (必填)— 重排序模型:
rerank.model、bge-reranker-v2-m3或cohere-rerank-3.5pinecone-rerank-v0 - (必填)— 用于重排序的字段
rerank.rankFields - (可选)— 重排序后最终返回的结果数量
rerank.topN
rerank-documents
rerank-documentsrerank-documents
rerank-documentsRerank a set of documents or records against a query without performing a vector search first.
Parameters:
- (required) —
model,bge-reranker-v2-m3, orcohere-rerank-3.5pinecone-rerank-v0 - (required) — The query to rerank against
query - (required) — Array of strings or records to rerank
documents - (required) — Number of results to return
options.topN - (optional) — If documents are records, the field(s) to rerank on
options.rankFields
在不先执行向量搜索的情况下,根据查询对一组文档或记录进行重排序。
参数:
- (必填)—
model、bge-reranker-v2-m3或cohere-rerank-3.5pinecone-rerank-v0 - (必填)— 用于重排序的查询
query - (必填)— 要重排序的字符串或记录数组
documents - (必填)— 返回的结果数量
options.topN - (可选)— 如果文档是记录,则指定用于重排序的字段
options.rankFields