Skip to content
Changelog

Report a bug

Pinecone integration: operations for AI agents

Pinecone connects to Type.com as an integration. This page lists the 169 operations documented for Claude and Codex in Type.com: 89 read, 51 write, 29 destructive.

About Pinecone

Long-term Memory for AI. The Pinecone vector database makes it easy to build high-performance vector search applications. Developer-friendly, fully managed, and easily scalable without infrastructure hassles.

Categories: Developer Tools & IT

How to connect Pinecone to Type.com

Access
Organization connection: one shared account for the workspace.

To add the connection, follow the steps in Space connections.

Documented operations

169 operations. Your connection's permissions determine which operations are available. API references describe the vendor surface; they do not expand your connection's credentials or access.

169 operations listed.

Manage Indexes

  • Read

    List indexes

    List all indexes in a project.

  • Read

    Describe an index

    Get a description of an index.

  • Read

    List backups for an index

    When includedeleted is false (or omitted), indexname must resolve to an active index in the project. If no active index by that name exists—including the case where only deleted indexes have used the name—the API returns 404, not an empty list.

  • Read

    List backup schedules for an index

    Return backup schedules associated with the given index.

  • Read

    List backups for all indexes in a project

    List all backups for a project.

  • Read

    Describe a backup

    Get a description of a backup.

  • Read

    Describe a backup schedule

    Get a single backup schedule by ID.

  • Read

    List backup history for a schedule

    Return backups that were created by the specified backup schedule.

  • Read

    List restore jobs

    List all restore jobs for a project.

  • Read

    Describe a restore job

    Get a description of a restore job.

  • Read

    List collections

    List all collections in a project. Collections are supported only for pod-based indexes. Pod-based indexes cannot be created with API version 2026-07, so this operation applies to pod-based indexes created with an earlier API version.

  • Read

    Describe a collection

    Get a description of a collection. Collections are supported only for pod-based indexes. Pod-based indexes cannot be created with API version 2026-07, so this operation applies to pod-based indexes created with an earlier API version.

  • Write

    Create an index

    Create a Pinecone index. Define the schema for your index — dense vector, sparse vector, and full-text search fields, and string fields with integrated embedding — and, optionally, the deployment infrastructure (managed serverless or BYOC).

  • Write

    Configure an index

    Configure an existing index. For guidance and examples, see Manage indexes.

  • Write

    Create a backup of an index

    Create a backup of an index. The backup records the schema of the index it is taken from and reports it as schema. An index created from the backup with Create index from backup inherits that schema.

  • Write

    Create a backup schedule for an index

    Create a time-based backup schedule for the specified index. Backup schedules are supported for serverless and BYOC indexes; the request is rejected with 400 for a pod-based index.

  • Write

    Create a legacy integrated index

    Create a legacy integrated index. Legacy integrated indexes created via create-for-model require the legacy Records API.

  • Write

    Create an index from a backup

    Create an index from a backup. The restored index inherits the schema of the index the backup was taken from, including its full-text search fields and its embedding configuration (embed / sparseembed fields, or the semantictext field of a legacy integrated index created via cre…

  • Write

    Update a backup schedule

    Update frequency, retention, or enabled state for a backup schedule. Re-enabling a disabled schedule (enabled: true) enqueues a new backup operation, and is rejected with 409 when the index already has another enabled schedule.

  • Write

    Create a collection

    Create a Pinecone collection. Collections are supported only for pod-based indexes. Pod-based indexes cannot be created with API version 2026-07, so this operation applies to pod-based indexes created with an earlier API version.

  • Destructive

    Delete an index

    Delete an existing index.

  • Destructive

    Delete a backup

    Delete a backup.

  • Destructive

    Delete a backup schedule

    Permanently remove a backup schedule.

  • Destructive

    Delete a collection

    Delete an existing collection. Collections are supported only for pod-based indexes. Pod-based indexes cannot be created with API version 2026-07, so this operation applies to pod-based indexes created with an earlier API version.

Bulk Operations

  • Read

    List imports

    List all recent and ongoing import operations. By default, listimports returns up to 100 imports per page. If the limit parameter is set, list returns up to that number of imports instead.

  • Read

    Describe an import

    Return details of a specific import operation. For guidance and examples, see Import data.

  • Write

    Start import

    Start an asynchronous import of vectors from object storage into an index. For guidance and examples, see Import data.

  • Destructive

    Cancel an import

    Cancel an import operation if it is not yet finished. It has no effect if the operation is already finished. For guidance and examples, see Import data.

Vector Operations

  • Read

    Get index stats

    Return statistics about the contents of an index, including the vector count per namespace, the number of dimensions, and the index fullness. Serverless indexes scale automatically as needed, so index fullness is relevant only for pod-based indexes.

  • Read

    Search with a vector

    Search a namespace using a query vector. It retrieves the ids of the most similar items in a namespace, along with their similarity scores. For guidance, examples, and limits, see Search.

  • Read

    Fetch records

    Look up and return records by ID from a single namespace. The returned records include the vector data and/or metadata. For on-demand indexes, since vector values are retrieved from object storage, fetch operations may have increased latency.

  • Read

    Fetch records by metadata

    Look up and return records by metadata from a single namespace. The returned records include the vector data and metadata. For guidance and examples, see Fetch data.

  • Read

    List record IDs

    List the IDs of records in a single namespace of a serverless index. An optional prefix can be passed to limit the results to IDs with a common prefix. Returns up to 100 IDs at a time by default in sorted order (bitwise "C" collation).

  • Read

    Search with text (legacy Records API)

    Search a namespace with a query text, query vector, or record ID and return the most similar records, along with their similarity scores. Optionally, rerank the initial results based on their relevance to the query.

  • Write

    Update a record

    Update records in a namespace, either by ID or with a metadata filter. By ID, you can change a single record's vector values, its metadata, or both. A filter updates metadata only, and applies to every record that matches.

  • Write

    Upsert records

    Upsert records into a namespace. If a new value is upserted for an existing record ID, it will overwrite the previous value. For guidance, examples, and limits, see Upsert data.

  • Write

    Upsert text (legacy Records API)

    Upsert text into a namespace. Pinecone converts the text to vectors automatically using the hosted embedding model associated with the index. This operation is part of the legacy Records API, which legacy integrated indexes require.

  • Destructive

    Delete records

    Delete records by id or by metadata from a single namespace. For guidance and examples, see Delete data.

Namespace Operations

  • Read

    List namespaces

    List all namespaces in a serverless index. Up to 100 namespaces are returned at a time by default, in sorted order (bitwise “C” collation). If the limit parameter is set, up to that number of namespaces are returned instead.

  • Read

    Describe a namespace

    Describe a namespace in a serverless index, including the total number of vectors in the namespace. For guidance and examples, see Manage namespaces. Note: This operation is not supported for pod-based indexes.

  • Write

    Create a namespace

    Create a namespace in a serverless index. For guidance and examples, see Manage namespaces. Note: This operation is not supported for pod-based indexes.

  • Destructive

    Delete a namespace

    Delete a namespace from a serverless index. Deleting a namespace is irreversible; all data in the namespace is permanently deleted. For guidance and examples, see Manage namespaces. Note: This operation is not supported for pod-based indexes.

Namespace Alias Operations

  • Read

    List namespace aliases

    List the namespace aliases in a serverless index, ordered by name. Filter by name prefix or by target namespace. Up to 100 aliases are returned per page; set limit to return fewer.

  • Read

    Describe a namespace alias

    Get an alias and the namespace it resolves to. For guidance and examples, see Manage namespace aliases.

  • Write

    Create a namespace alias

    Create an alias that resolves to an existing namespace in the index. Reads through the alias are served from its target namespace; writes to an alias name are rejected. Many aliases can resolve to the same namespace.

  • Write

    Repoint a namespace alias

    Update the namespace an alias resolves to. The change is atomic: the target is validated in the same transaction that rewrites the alias, and on failure the alias keeps its current target.

  • Destructive

    Delete a namespace alias

    Remove an alias. The target namespace and its data are unchanged. Reads that still use the alias name are treated as reads of a nonexistent namespace (empty results, not an error), so migrate readers off the alias first.

Document Operations

  • Read

    Search documents

    Search for documents in a namespace using one or more scoring methods (dense vector, sparse vector, text, or query string similarity, or similarity to query text that Pinecone embeds with a field's own model).

  • Read

    Fetch documents

    Fetch documents from a namespace. Returns the specified fields for each document. Exactly one of ids or filter must be specified. - ids: Fetch the documents with the given IDs. - filter: Fetch every document matching a metadata filter expression.

  • Read

    List documents

    List documents in a namespace. Returns up to 100 documents per page by default, in sorted order (bitwise "C" collation). Use the optional prefix parameter to limit results to documents whose IDs start with a given prefix.

  • Write

    Upsert documents

    Upsert documents into a namespace. Each document must include an id field and at least one field defined in the index schema; metadata fields may be provided alongside them.

  • Write

    Update documents

    Apply partial updates to documents in a namespace. Documents are selected either per ID with documents, or in bulk with filter. - documents: Each update is identified by its id.

  • Destructive

    Delete documents

    Delete documents from a namespace. Exactly one of ids, filter, or deleteall must be specified. - ids: Delete documents with the given IDs. - filter: Delete every document matching a metadata filter expression.

Inference

  • Read

    Describe a model

    Get a description of a model hosted by Pinecone. You can use hosted models for integrated inference in Pinecone operations, such as integrated-embedding fields in a document index, or for standalone embedding and reranking.

  • Write

    Generate vectors

    Generate vector embeddings for input data. This endpoint uses Pinecone's hosted embedding models.

  • Write

    Rerank results

    Rerank results according to their relevance to a query. For guidance and examples, see Rerank results.

Query

  • Read

    List the selectable query models

    What the model field on POST /query draws from: the default, every catalog entry, the tier → model-id map, per-phase defaults, and the curate-capable ids. Only entries with available: true are selectable; the rest are rejected with 400.

  • Read

    Fetch one query (turn)

    One turn, including its steps, citations, and usage. This is what a background run is polled with.

  • Read

    Fetch the recorded trace for a query turn

    The per-turn debug trace (steps, tool calls, strategy, cost, rollup). Trace persistence is unconditional — every turn lands a trace blob — so this is available once the turn is terminal.

  • Read

    Stream a query turn's events (SSE)

    The turn's events as they land — the same type-named events POST /query emits with stream: true, as a standalone resumable subscription. Reconnect with Last-Event-ID to replay from where the stream dropped.

  • Read

    List project-wide query sessions (newest first)

    Every conversation thread in the project, newest first.

  • Read

    Get a session with its queries (conversation order)

    One session with its turns in conversation order.

  • Write

    Run one KnowQL query turn

    Send ask and, on a new session, a scope of 1–10 contexts; continue an existing one with sessionid or previousqueryid. Scoped search contexts must be curated, and a scope may not mix work and search contexts. Read the answer from output[].content[].text.

  • Write

    Record thumbs-up/down feedback on a query turn

    Record a thumbs-up or thumbs-down on a turn. The rating is returned on the turn thereafter.

  • Destructive

    Cancel an in-progress query turn

    Idempotent — an already-terminal query is returned unchanged. On success the turn's status becomes cancelled.

  • Destructive

    Delete a session and its queries

    Delete a session and every turn belonging to it.

Projects

  • Read

    List projects

    List all projects in an organization.

  • Read

    Get project details

    Get a project's details.

  • Write

    Create a new project

    Create a new project.

  • Write

    Update a project

    Update a project's name, maximum number of Pods, or customer-managed encryption key (CMEK).

  • Destructive

    Delete a project

    Delete a project and all its configuration; delete its indexes, assistants, backups, and collections first.

Organizations

  • Read

    List organizations

    List all organizations associated with an account.

  • Read

    Get organization details

    Get an organization's details.

  • Write

    Update an organization

    Update an organization's name.

  • Destructive

    Delete an organization

    Delete an organization and all its configuration; delete all its projects first.

API Keys

  • Read

    List API keys

    List all API keys in a project.

  • Read

    Get API key details

    Get an API key's details, excluding its secret.

  • Write

    Create an API key

    Create an API key for a project to authenticate Data Plane and Control Plane requests.

  • Write

    Update an API key

    Update an API key's name and roles.

  • Destructive

    Delete an API key

    Delete an API key from a project.

Users

  • Read

    List users in the organization

    List users in the caller's organization, optionally filtered by email address.

  • Read

    Get user details

    Get a user in the caller's organization by ID.

  • Destructive

    Remove a user from the organization

    Remove a user from the organization and revoke their role bindings; their Pinecone account is not deleted.

Invites

  • Read

    List invites

    List pending and expired invites in the caller's organization.

  • Read

    Get invite details

    Get an invite in the caller's organization by ID.

  • Write

    Invite a user to the organization

    Invite a user to the organization by email and grant their initial role bindings.

  • Write

    Resend an invite email

    Resend the invite email and extend its expiration to 7 days from now; limited to 100 emails per hour per organization.

  • Destructive

    Delete an invite

    Delete a pending or expired invite and its role bindings; to remove an accepted user, delete the user instead.

Service Accounts

  • Read

    List service accounts

    List service accounts in the caller's organization.

  • Read

    Get service account details

    Get a service account by ID; the client secret is returned only from create and rotate-secret requests.

  • Write

    Create a service account

    Create a service account with optional initial role bindings; the client secret is returned only once.

  • Write

    Update a service account

    Update a service account's name; role bindings are managed through the role-binding endpoints.

  • Write

    Rotate a service account's OAuth client secret

    Rotate a service account's OAuth client secret; the previous secret and its tokens are revoked within seconds and the new secret is returned only once.

  • Destructive

    Delete a service account

    Delete a service account and its role bindings; tokens it minted are revoked within a few seconds.

Role Bindings

  • Read

    List role bindings

    List role bindings in the caller's organization, optionally filtered by principal, resource, and role.

  • Read

    Get role binding details

    Get a role binding in the caller's organization by ID.

  • Write

    Create a role binding

    Grant a role to a principal at an organization or project scope.

  • Destructive

    Delete a role binding

    Delete a role binding; permissions are revoked when the deletion completes.

Cost Usage Exports

  • Read

    List cost usage exports

    List the cost usage exports belonging to your organization, newest first. List items never carry a download link: downloadurl and downloadurlexpiresat are always null here. Request a single export to generate one.

  • Read

    Get cost usage export details

    Get the state of a cost usage export, and a link to download its file once the export has completed. A fresh signed link, valid for one hour, is generated on each read.

  • Write

    Create a cost usage export

    Start an asynchronous export of an organization's cost usage data. Only one export can run at a time and only a limited number of exports can be triggered per hour. The generated file is kept for two days.

Manage Assistants

  • Read

    List assistants

    List of all assistants in a project. For guidance and examples, see Manage assistants.

  • Read

    Check assistant status

    Get the status of an assistant. For guidance and examples, see Manage assistants

  • Read

    List Files

    List all files in an assistant, with an option to filter files with metadata. For guidance and examples, see Manage files.

  • Read

    Describe a file

    Get the current status and metadata of a file uploaded to an assistant.

  • Read

    List operations

    List all operations for an assistant. Returns operations that are in progress, as well as recently completed or failed operations. Both successful and failed operations are retained for 30 days after completion. Use the operationtype and status query parameters to filter results.

  • Read

    Describe an operation

    Get the status of an operation.

  • Read

    Retrieve context from an assistant

    Retrieve context snippets from an assistant to use as part of RAG or any agentic flow. For guidance and examples, see Retrieve context snippets.

  • Write

    Create an assistant

    Create an assistant. This is where you specify the underlying training model, which cloud provider you would like to deploy with, and more. For guidance and examples, see Create an assistant

  • Write

    Update an assistant

    Update an existing assistant. You can modify the assistant's instructions. For guidance and examples, see Manage assistants.

  • Write

    Upload a file

    Upload a file to the specified assistant. An identifier will be generated. To specify a file identifier or to replace file content, use the upsert endpoint (PUT /files/{assistantname}/{assistantfileid}). This operation is asynchronous.

  • Write

    Chat through an OpenAI-compatible interface

    Chat with an assistant. This endpoint is based on the OpenAI Chat Completion API, a commonly used and adopted API. It is useful if you need inline citations or OpenAI-compatible responses, but has limited functionality compared to the standard chat interface.

  • Write

    Chat with an assistant

    Chat with an assistant and get back citations in structured form. This is the recommended way to chat with an assistant, as it offers more functionality and control over the assistant's responses and references than the OpenAI-compatible chat interface.

  • Destructive

    Delete an assistant

    Delete an existing assistant. For guidance and examples, see Manage assistants

  • Destructive

    Upsert a file

    Create or replace a file in the specified assistant. If a file with the given assistantfileid already exists, it will be replaced with the new file. If it doesn't exist, a new file will be created with that identifier. This operation is asynchronous.

  • Destructive

    Delete a file

    Delete an uploaded file from an assistant. This operation is asynchronous. The response includes an operation ID that can be used to poll for completion via the describe operation endpoint.

Metrics

  • Read

    Prometheus HTTP Service Discovery

    Get endpoints for Prometheus scraping.

  • Write

    Evaluate an answer

    Evaluate the correctness and completeness of a response from an assistant or a RAG system. The correctness and completeness are evaluated based on the precision and recall of the generated answer with respect to the ground truth answer facts.

Manage Workspaces

  • Read

    List workspaces

    List the workspaces in the project. Results are paginated; the pagination object is omitted when there are no further results.

  • Read

    Describe a workspace

    Get a description of a workspace.

  • Write

    Create a workspace

    Create a Nexus workspace. The workspace name must be unique within the project. Workspaces are created asynchronously. On success the workspace is returned in the Initializing state with ready set to false.

  • Destructive

    Delete a workspace

    Delete an existing workspace. Deletion is asynchronous: the workspace transitions to the Terminating state and its contexts are deleted, after which the workspace itself is removed.

Auth

  • Read

    Current identity (whoami)

    Who the current credential authenticates, and which project and workspace it is scoped to.

  • Write

    Exchange a Pinecone API key for a session token

    Exchange a Pinecone API key for the session token every other endpoint expects. Send the token as a bearer credential.

Project

  • Read

    Get the active project

    The project the session is scoped to, plus whether this deployment requires a workspace host.

  • Read

    Project-wide task stats

    Aggregate task counters across the whole project (all contexts), broken down by state and workflow. The per-context equivalent is GET /contexts/{slug}/tasks/stats.

Connectors

  • Read

    List the project's linked connectors

    The project's linked source providers. Filter by kind to those valid for a context kind.

  • Read

    Get a connector

    One linked connector, secrets stripped.

  • Read

    Start the OAuth link flow for a provider

    Returns an authorizeurl for the user to open in a browser. The connector is persisted by the provider's OAuth redirect callback once the user approves access.

  • Read

    Browse a connector's folders/files

    Browse the provider's folders and files to find the ids to import. Omit folder for the root.

  • Read

    List a context's connectors (with per-context enabled flag)

    The project's connectors, each with whether this context may import from it.

  • Write

    Link a connector with an API key

    Link a key-based provider by supplying its API key directly, instead of running the OAuth flow.

  • Destructive

    Unlink a connector

    Unlink the connector from the project. Sources it already imported stay in place.

  • Destructive

    Enable or disable a connector for a context

    Enable or disable one connector for this context.

Contexts

  • Read

    List contexts (newest first, with stats)

    Every context in the active project, newest first, each with its aggregate task counters.

  • Read

    Get a context

    One context and its derived lifecycle flags.

  • Read

    Per-context task statistics

    Aggregate counters over every task this context has run. The project-wide equivalent is GET /stats.

  • Write

    Create a context

    Created empty, and not queryable until you import sources and curate them — curate is explicit. Optionally seed a manifest. A work context is the exception: queryable from day zero, built from traces of work rather than documents.

  • Write

    Trigger an on-demand self-tuning optimize

    Tunes the manifest from real query traffic, then chains a forced re-curate. Every field of the required body is optional, so {} is valid — and a no-op without candidatequeries. The context must be curated, and only one optimize runs at a time.

  • Write

    Run the Design-flow explore agent (propose a manifest)

    Inspects the uploaded source and proposes manifest-template matches in the task's output (see ExploreOutput). Persists nothing — review the proposal, then apply it with a manifest update and a forced curate. Body is optional.

  • Write

    Estimate the token + time cost of curating the in-progress manifest

    Estimates the token and time cost of curating the sources under the manifest in the request body, reported as the task's output (see ProfileEstimateOutput). Persists nothing. Body is optional.

  • Destructive

    Update a context

    All fields optional; an absent one is left untouched. An empty string clears description/guide, and {} clears the manifest back to defaults.

  • Destructive

    Delete a context (and its Pinecone indexes)

    Delete the context, its sources, its curated knowledge, and its Pinecone indexes. Not reversible.

Manifest

  • Read

    Get the context's manifest

    The context's pinned manifest document. A context that pins nothing returns null — the runtime fills defaults. Writes go through the manifest field on PUT /contexts/{slug}.

  • Read

    List the built-in manifest templates

    The templates a new context can be seeded from.

Curation

  • Read

    Get the curation ledger

    What the last curate recorded: the per-source hash and timestamp map, artifact edges, corpus groups, the live version pin, delete tombstones, artifact-reclaim intents, and the pointer and glossary resume cursors.

  • Write

    Build or rebuild the context's index from its sources

    Curate the staged sources under the active manifest. Required before querying — there is no auto-curate. Body is optional; force: true rebuilds fully, which is what a manifest edit needs. Search contexts only: a work context builds via work/groom.

  • Write

    Drop named sources from the index + ledger

    Drop named sources from the index and the curation ledger without re-reading the rest of the corpus.

Source Files

  • Read

    List the source file tree (root)

    The root of the source tree — what has been staged but not necessarily curated.

  • Read

    Source file/dir/size counts

    Aggregate counts. Where source limits apply, five more fields carry the caps and usage against them (maxfilespercontext, maxbytespercontext, maxbytesperfile, usedfiles, usedbytes); they are absent otherwise.

  • Read

    Flat manifest of source files (path, size, lastmodified)

    Flat source-object inventory. Excludes the inbox/ staging directory.

  • Read

    List the source file tree under a path

    One directory of the source tree.

  • Read

    Read a source file (raw bytes)

    The raw bytes of one staged source file.

  • Write

    Import source documents from a project connector

    Pull source documents from a linked project connector (Box, ...) into the context. Supply either folder or files. Stages the source only — curate explicitly before querying.

  • Write

    Upload a source file or archive

    One file per request, max 2 GiB; archives (.zip, .tar, .tar.gz, .tgz) are expanded by the import runtime. Stages the source without indexing it — curate explicitly before querying.

  • Destructive

    Delete a source file

    Remove one file from the source tree. Its index entries survive until the next curate.

Knowledge

  • Read

    List curated knowledge (root)

    The root of the curated knowledge tree — chunks and artifacts the last curate wrote.

  • Read

    Knowledge file/dir/size counts

    File, directory, and byte counts over the curated knowledge tree.

  • Read

    List curated knowledge under a path

    One directory of the curated knowledge tree.

  • Read

    Read a curated knowledge file (raw bytes)

    The raw bytes of one curated knowledge file.

Tasks

  • Read

    List tasks (project-wide, paginated)

    Task records across the project, newest first. Filter by workflow, state, or context, and page with limit plus page or offset.

  • Read

    Get a task (with steps)

    Unlike the listing, this carries the task's steps — every one by default, or the newest N with stepslimit, where stepstotal reports the true count.

  • Destructive

    Cancel/terminate a task

    Body is optional. Max body 1 MiB.

Task Files

  • Read

    Read a task file (raw bytes)

    Serves a task's files while it runs and after it finishes. This is how a finished pack's .context.zip is downloaded.

  • Read

    Stat a task file

    Stat one file in the task container without reading its bytes.

  • Read

    List a task's files under a path

    One directory inside the task's working directory.

  • Read

    List a task's working-directory files

    The root of the task's working directory.

  • Destructive

    Delete a task file

    Delete one file from a running task container.

OAuth

  • Write

    Create an access token

    Obtain an access token for a service account using the OAuth2 client credentials flow. An access token is needed to authorize requests to the Pinecone Admin API. The host domain for OAuth endpoints is login.pinecone.io.

See How connections work for access and permissions, or browse all integrations.