mcp

Name: mcp
Author: pinecone-io

23stars1forks

How to install mcp?

pinecone-io/gemini-cli-extension/mcp

$npx -y skills add pinecone-io/gemini-cli-extension --skill mcp

Installs into the current project.

Summary

Reference for the Pinecone MCP server tools. Documents all available tools - list-indexes, describe-index, describe-index-stats, create-index-for-model, upsert-records, search-records, cascading-search, and rerank-documents. Use when an agent needs to understand what Pinecone MCP

Files · 1

View on GitHub

SKILL.md

1	# Pinecone MCP Tools Reference
2
3	The Pinecone MCP server exposes the following tools to AI agents and IDEs. For setup and installation instructions, see the [MCP server guide](https://docs.pinecone.io/guides/operations/mcp-server#tools).
4
5	> 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.
6
7	---
8
9	## `list-indexes`
10
11	List all indexes in the current Pinecone project.
12
13	---
14
15	## `describe-index`
16
17	Get configuration details for a specific index — cloud, region, dimension, metric, embedding model, field map, and status.
18
19	Parameters:
20	- `name` (required) — Index name
21
22	---
23
24	## `describe-index-stats`
25
26	Get statistics for an index including total record count and per-namespace breakdown.
27
28	Parameters:
29	- `name` (required) — Index name
30
31	---
32
33	## `create-index-for-model`
34
35	Create a new serverless index with an integrated embedding model. Pinecone handles embedding automatically — no external model needed.
36
37	Parameters:
38	- `name` (required) — Index name
39	- `cloud` (required) — `aws`, `gcp`, or `azure`
40	- `region` (required) — Cloud region (e.g. `us-east-1`)
41	- `embed.model` (required) — Embedding model: `llama-text-embed-v2`, `multilingual-e5-large`, or `pinecone-sparse-english-v0`
42	- `embed.fieldMap.text` (required) — The record field that contains text to embed (e.g. `chunk_text`)
43
44	---
45
46	## `upsert-records`
47
48	Insert or update records in an integrated index. Records are automatically embedded using the index's configured model.
49
50	Parameters:
51	- `name` (required) — Index name
52	- `namespace` (required) — Namespace to upsert into
53	- `records` (required) — Array of records. Each record must have an `id` or `_id` field and contain the text field specified in the index's `fieldMap`. Do not nest fields under `metadata` — put them directly on the record.
54
55	Example record:
56	```json
57	{ "_id": "rec1", "chunk_text": "The Eiffel Tower was built in 1889.", "category": "architecture" }
58	```
59
60	---
61
62	## `search-records`
63
64	Semantic text search against an integrated index. Pass plain text — the MCP embeds the query automatically using the index's model.
65
66	Parameters:
67	- `name` (required) — Index name
68	- `namespace` (required) — Namespace to search
69	- `query.inputs.text` (required) — The text query
70	- `query.topK` (required) — Number of results to return
71	- `query.filter` (optional) — Metadata filter using MongoDB-style operators (`$eq`, `$ne`, `$in`, `$gt`, `$gte`, `$lt`, `$lte`)
72	- `rerank.model` (optional) — Reranking model: `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
73	- `rerank.rankFields` (optional) — Fields to rerank on (e.g. `["chunk_text"]`)
74	- `rerank.topN` (optional) — Number of results to return after reranking
75
76	---
77
78	## `cascading-search`
79
80	Search across multiple indexes simultaneously, then deduplicate and rerank results into a single ranked list.
81
82	Parameters:
83	- `indexes` (required) — Array of `{ name, namespace }` objects to search across
84	- `query.inputs.text` (required) — The text query
85	- `query.topK` (required) — Number of results to retrieve per index before reranking
86	- `rerank.model` (required) — Reranking model: `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
87	- `rerank.rankFields` (required) — Fields to rerank on
88	- `rerank.topN` (optional) — Final number of results to return after reranking
89
90	---
91
92	## `rerank-documents`
93
94	Rerank a set of documents or records against a query without performing a vector search first.
95
96	Parameters:
97	- `model` (required) — `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
98	- `query` (required) — The query to rerank against
99	- `documents` (required) — Array of strings or records to rerank
100	- `options.topN` (required) — Number of results to return
101	- `options.rankFields` (optional) — If documents are records, the field(s) to rerank on

People who liked this also viewed

6 picks

Type

mcp

23stars1forks

View on GitHub

How to install mcp?

pinecone-io/gemini-cli-extension/mcp

$npx -y skills add pinecone-io/gemini-cli-extension --skill mcp

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Pinecone MCP Tools Reference
2
3	The Pinecone MCP server exposes the following tools to AI agents and IDEs. For setup and installation instructions, see the [MCP server guide](https://docs.pinecone.io/guides/operations/mcp-server#tools).
4
5	> 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.
6
7	---
8
9	## `list-indexes`
10
11	List all indexes in the current Pinecone project.
12
13	---
14
15	## `describe-index`
16
17	Get configuration details for a specific index — cloud, region, dimension, metric, embedding model, field map, and status.
18
19	Parameters:
20	- `name` (required) — Index name
21
22	---
23
24	## `describe-index-stats`
25
26	Get statistics for an index including total record count and per-namespace breakdown.
27
28	Parameters:
29	- `name` (required) — Index name
30
31	---
32
33	## `create-index-for-model`
34
35	Create a new serverless index with an integrated embedding model. Pinecone handles embedding automatically — no external model needed.
36
37	Parameters:
38	- `name` (required) — Index name
39	- `cloud` (required) — `aws`, `gcp`, or `azure`
40	- `region` (required) — Cloud region (e.g. `us-east-1`)
41	- `embed.model` (required) — Embedding model: `llama-text-embed-v2`, `multilingual-e5-large`, or `pinecone-sparse-english-v0`
42	- `embed.fieldMap.text` (required) — The record field that contains text to embed (e.g. `chunk_text`)
43
44	---
45
46	## `upsert-records`
47
48	Insert or update records in an integrated index. Records are automatically embedded using the index's configured model.
49
50	Parameters:
51	- `name` (required) — Index name
52	- `namespace` (required) — Namespace to upsert into
53	- `records` (required) — Array of records. Each record must have an `id` or `_id` field and contain the text field specified in the index's `fieldMap`. Do not nest fields under `metadata` — put them directly on the record.
54
55	Example record:
56	```json
57	{ "_id": "rec1", "chunk_text": "The Eiffel Tower was built in 1889.", "category": "architecture" }
58	```
59
60	---
61
62	## `search-records`
63
64	Semantic text search against an integrated index. Pass plain text — the MCP embeds the query automatically using the index's model.
65
66	Parameters:
67	- `name` (required) — Index name
68	- `namespace` (required) — Namespace to search
69	- `query.inputs.text` (required) — The text query
70	- `query.topK` (required) — Number of results to return
71	- `query.filter` (optional) — Metadata filter using MongoDB-style operators (`$eq`, `$ne`, `$in`, `$gt`, `$gte`, `$lt`, `$lte`)
72	- `rerank.model` (optional) — Reranking model: `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
73	- `rerank.rankFields` (optional) — Fields to rerank on (e.g. `["chunk_text"]`)
74	- `rerank.topN` (optional) — Number of results to return after reranking
75
76	---
77
78	## `cascading-search`
79
80	Search across multiple indexes simultaneously, then deduplicate and rerank results into a single ranked list.
81
82	Parameters:
83	- `indexes` (required) — Array of `{ name, namespace }` objects to search across
84	- `query.inputs.text` (required) — The text query
85	- `query.topK` (required) — Number of results to retrieve per index before reranking
86	- `rerank.model` (required) — Reranking model: `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
87	- `rerank.rankFields` (required) — Fields to rerank on
88	- `rerank.topN` (optional) — Final number of results to return after reranking
89
90	---
91
92	## `rerank-documents`
93
94	Rerank a set of documents or records against a query without performing a vector search first.
95
96	Parameters:
97	- `model` (required) — `bge-reranker-v2-m3`, `cohere-rerank-3.5`, or `pinecone-rerank-v0`
98	- `query` (required) — The query to rerank against
99	- `documents` (required) — Array of strings or records to rerank
100	- `options.topN` (required) — Number of results to return
101	- `options.rankFields` (optional) — If documents are records, the field(s) to rerank on

People who liked this also viewed

6 picks

Type