Verfy

MCP servers / @pinecone-database/mcp

MCP tool changes

@pinecone-database/mcp

3 of its last 5 updates changed what the agent reads: 13 tool descriptions rewritten, up to 0.4.0.

npm package · last recorded on 13 November 2026 on a clean machine with placeholder settings; only tools/list was sent. A changed description is not a vulnerability. npm page · Method and limits · this data as JSON.

Release by release

0.3.0 → 0.4.0

minor release, 29 October 2026

Added: create-document-index, search-document-index, upsert-documents

create-index-for-model: description changed
Before (0.3.0)
Create a Pinecone index with integrated inference, which automatically embeds the text field named in "fieldMap". Supports AWS, GCP, and Azure cloud providers. This call blocks until the index is ready to accept upserts and queries, so no readiness polling is needed afterwards. If an index with the same name already exists, it is returned unchanged instead of erroring.
After (0.4.0)
Create a Pinecone index with integrated inference, which automatically embeds the text field named in "fieldMap". Supports AWS, GCP, and Azure cloud providers. This call blocks until the index is ready to accept upserts and queries, so no readiness polling is needed afterwards. If an index with the same name already exists, it is returned unchanged instead of erroring. For a new index, prefer create-document-index, which supports full-text search and integrated embedding on the same field.
search-records: description changed
Before (0.3.0)
Search an index for records that are semantically similar to the query text. Only works with integrated-inference indexes (e.g. created with create-index-for-model), which embed the query automatically; it fails on standard indexes — use describe-index to check the index type. Returns hits with "_id", "_score", and the record's stored "fields". Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.
After (0.4.0)
Search an index for records that are semantically similar to the query text. Only works with integrated-inference indexes (e.g. created with create-index-for-model), which embed the query automatically; it fails on standard indexes and document indexes (use search-document-index for those) — use describe-index to check the index type. Returns hits with "_id", "_score", and the record's stored "fields". Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.
upsert-records: description changed
Before (0.3.0)
Insert or update records in an integrated Pinecone index. Text in the field named by the index's "fieldMap" is embedded automatically — call describe-index first to confirm that field name. Records with an existing "id" are overwritten. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.
After (0.4.0)
Insert or update records in an integrated Pinecone index. For a document index, use upsert-documents instead. Text in the field named by the index's "fieldMap" is embedded automatically — call describe-index first to confirm that field name. Records with an existing "id" are overwritten. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.

0.2.1 → 0.3.0

minor release, 7 August 2026
cascading-search: description and input schema changed
Before (0.2.1)
Search across multiple indexes for records that are similar to the query text, deduplicate and rerank the results.
After (0.3.0)
Search across multiple indexes for records that are similar to the query text, then deduplicate and rerank the combined results. Reranking is required for this tool. Only works with integrated-inference indexes. To search a single index, use search-records instead.
create-index-for-model: description changed
Before (0.2.1)
Create a Pinecone index with integrated inference. Supports AWS, GCP, and Azure cloud providers.
After (0.3.0)
Create a Pinecone index with integrated inference, which automatically embeds the text field named in "fieldMap". Supports AWS, GCP, and Azure cloud providers. This call blocks until the index is ready to accept upserts and queries, so no readiness polling is needed afterwards. If an index with the same name already exists, it is returned unchanged instead of erroring.
describe-index: description changed
Before (0.2.1)
Describe the configuration of a Pinecone index
After (0.3.0)
Describe the configuration of a Pinecone index, including its embedding model and the "fieldMap" that names the record field holding the text to embed. Call this before upsert-records to learn the required field name. Check "status.ready" in the response to confirm the index can accept upserts and queries.
describe-index-stats: description changed
Before (0.2.1)
Describe the statistics of a Pinecone index and its namespaces
After (0.3.0)
Describe the statistics of a Pinecone index, including record counts per namespace. Use this to discover which namespaces exist in an index. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.
list-indexes: description changed
Before (0.2.1)
List all Pinecone indexes
After (0.3.0)
List all Pinecone indexes in the project, including each index's configuration and readiness status. Use this to discover valid index names before calling other tools.
rerank-documents: description and input schema changed
Before (0.2.1)
Rerank a set of documents based on a query
After (0.3.0)
Rerank a set of provided documents by relevance to a query. Use this for documents that did not come from a Pinecone search (e.g. externally sourced text). To rerank the results of a single-index search, prefer the "rerank" parameter of search-records instead of calling this tool separately.
search-docs: description changed
Before (0.2.1)
Search Pinecone documentation for relevant information
After (0.3.0)
Search the official Pinecone documentation for relevant information. Use this before writing Pinecone code or explaining Pinecone behavior, rather than relying on assumptions.
search-records: description and input schema changed
Before (0.2.1)
Search an index for records that are similar to the query text
After (0.3.0)
Search an index for records that are semantically similar to the query text. Only works with integrated-inference indexes (e.g. created with create-index-for-model), which embed the query automatically; it fails on standard indexes — use describe-index to check the index type. Returns hits with "_id", "_score", and the record's stored "fields". Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.
upsert-records: description and input schema changed
Before (0.2.1)
Insert or update records in a Pinecone index
After (0.3.0)
Insert or update records in an integrated Pinecone index. Text in the field named by the index's "fieldMap" is embedded automatically — call describe-index first to confirm that field name. Records with an existing "id" are overwritten. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.

0.2.0 → 0.2.1

patch release, 5 February 2026

No tool definition changed.

0.1.17 → 0.2.0

minor release, 27 January 2026

cascading-search: input schema changed; new arguments: llm_model, llm_provider

create-index-for-model: description and input schema changed; new arguments: cloud, llm_model, llm_provider, region
Before (0.1.17)
Create a Pinecone index with integrated inference
After (0.2.0)
Create a Pinecone index with integrated inference. Supports AWS, GCP, and Azure cloud providers.

describe-index: input schema changed; new arguments: llm_model, llm_provider

describe-index-stats: input schema changed; new arguments: llm_model, llm_provider

list-indexes: input schema changed; new arguments: llm_model, llm_provider

rerank-documents: input schema changed; new arguments: llm_model, llm_provider

search-docs: input schema changed

search-records: input schema changed; new arguments: llm_model, llm_provider

upsert-records: input schema changed; new arguments: llm_model, llm_provider

0.1.16 → 0.1.17

patch release, 25 September 2025

No tool definition changed.

Releases tried: 0.1.16 (2025-09-15); 0.1.17 (2025-09-25); 0.2.0 (2026-01-27); 0.2.1 (2026-02-05); 0.3.0 (2026-08-07); 0.4.0 (2026-10-29).

Tools in 0.4.0

The pin is the SHA-256 of the tool's name, description and input schema, as verfy.lock records it. The same pin in your lock means your agent reads this exact definition. Verfy 0.16.0 and later also pin a tool's title, annotations and output schema when the server sends them, so such a tool shows a different pin in your lock.

Tools in 0.4.0
ToolDescriptionPin
search-docsSearch the official Pinecone documentation for relevant information. Use this before writing Pinecone code or explaining Pinecone behavior, rather than relying on assumptions.05858ae99269
list-indexesList all Pinecone indexes in the project, including each index's configuration and readiness status. Use this to discover valid index names before calling other tools.139fcc436aa7
describe-indexDescribe the configuration of a Pinecone index, including its embedding model and the "fieldMap" that names the record field holding the text to embed. Call this before upsert-records to learn the required field name. Check "status.ready" in the response to confirm the index can accept upserts and queries.227665ce653d
describe-index-statsDescribe the statistics of a Pinecone index, including record counts per namespace. Use this to discover which namespaces exist in an index. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.368a44713b23
create-document-indexCreate a Pinecone document index: an index defined by a schema of searchable "string" fields. Each field can declare "fullTextSearch" (indexed for keyword search with BM25), "embed" (Pinecone embeds the text with a hosted dense model for semantic search), "sparseEmbed" (a hosted sparse model), or a combination. A good default is one field with both, for example {"chunk_text": {"fullTextSearch": {"language": "en"}, "embed": {}}}, which supports search by meaning and by exact terms on the same text. Fields you only need to filter on are not declared here; send them as document metadata. Use upsert-documents and search-document-index with the new index. This call blocks until the index is ready. If an index with the same name already exists, it is returned unchanged instead of erroring.54c02a9879ee
upsert-documentsInsert or update documents in a Pinecone document index (one created with create-document-index). Each document needs an "_id" and plain text in every field the index declares with "embed"; Pinecone embeds that text on write, so never send vectors. Call describe-index first to see the schema. Fields not in the schema are stored as metadata you can filter on. Documents with an existing "_id" are replaced. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.4d4d497da41b
search-document-indexSearch a Pinecone document index (one created with create-document-index). Call describe-index first to see which fields declare "embed", "sparseEmbed", or "fullTextSearch". Rank with one or more "scoreBy" clauses: - "embed": search by meaning on a field with "embed". Pass the query as plain text; Pinecone embeds it with the field's model. - "sparse_embed": search a field with "sparseEmbed" the same way. - "text": keyword (BM25) search on one or more full-text fields. Good for exact terms like IDs, error codes, and names. - "query_string": Lucene-style query syntax. Don't pass "fields" (the API rejects it); target fields with qualifiers inside the query, for example 'chunk_text:"exact phrase"', or omit qualifiers to search every full-text field. An "embed" or "sparse_embed" clause must be the only clause; "text" and "query_string" clauses can be combined. To rank by meaning and by exact terms, run two separate searches: their scores are on different scales, so don't compare them across searches. Returns matches with "_id", "_score", and the fields named in "includeFields". Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.7d74be9c17e8
create-index-for-modelCreate a Pinecone index with integrated inference, which automatically embeds the text field named in "fieldMap". Supports AWS, GCP, and Azure cloud providers. This call blocks until the index is ready to accept upserts and queries, so no readiness polling is needed afterwards. If an index with the same name already exists, it is returned unchanged instead of erroring. For a new index, prefer create-document-index, which supports full-text search and integrated embedding on the same field.db70f4c45976
upsert-recordsInsert or update records in an integrated Pinecone index. For a document index, use upsert-documents instead. Text in the field named by the index's "fieldMap" is embedded automatically — call describe-index first to confirm that field name. Records with an existing "id" are overwritten. Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.03281f56289c
search-recordsSearch an index for records that are semantically similar to the query text. Only works with integrated-inference indexes (e.g. created with create-index-for-model), which embed the query automatically; it fails on standard indexes and document indexes (use search-document-index for those) — use describe-index to check the index type. Returns hits with "_id", "_score", and the record's stored "fields". Newly upserted records can take a few seconds to be indexed and become visible to search results and index stats; if recent data is missing, wait briefly and retry.d57e3ca3a38e
rerank-documentsRerank a set of provided documents by relevance to a query. Use this for documents that did not come from a Pinecone search (e.g. externally sourced text). To rerank the results of a single-index search, prefer the "rerank" parameter of search-records instead of calling this tool separately.ed730240d768
cascading-searchSearch across multiple indexes for records that are similar to the query text, then deduplicate and rerank the combined results. Reranking is required for this tool. Only works with integrated-inference indexes. To search a single index, use search-records instead.377ad43dd3db

Lock these tools

Pin what @pinecone-database/mcp serves you today, commit the lock, and a pull request fails when a tool changes, with the text before and after. Pin the package version in your MCP config as well.

$ pip install verfy
$ verfy lock check --update   # shows each tool, then writes verfy.lock
$ git add verfy.lock
# .github/workflows/verfy.yml, after actions/checkout
- uses: verfy-dev/verfy-action@v1

Maintain this server? Run the same check on your releases and add the Locked by Verfy badge to your README.