Skip to content
Changelog

Report a bug

Algolia integration: operations for AI agents

Algolia connects to Type.com as an integration. This page lists the 208 operations documented for Claude and Codex in Type.com: 112 read, 39 write, 57 destructive.

About Algolia

Algolia REST and SDK helper references, subject to Type setup and account permissions.

Categories: Developer Tools & IT

How to connect Algolia to Type.com

Connect
Configure provider-specific targets and authentication setup.
Access
Organization connection: one shared account for the workspace.

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

Documented operations

208 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.

208 operations listed.

search

  • Read

    Search an index

    Searches a single index and returns matching search results as hits. This method lets you retrieve up to 1,000 hits. If you need more, use the [`browse` operation](https://www.algolia.com/doc/rest-api/search/browse) or increase the `paginatedLimitedTo` index setting.

    POST /1/indexes/{indexName}/query

  • Read

    Search multiple queries

    Runs multiple search queries against one or more indices in a single API request. Use cases include: - Searching different indices, such as products and marketing content. - Run multiple queries on the same index with different parameters or filters. If you know the expected result type, use the `searchForHits` or `searchForFacets` helper to simplify the response format.

    POST /1/indexes/*/queries

  • Read

    Search for facet values

    Searches for values of a specified facet attribute. - By default, facet values are sorted by decreasing count. You can adjust this with the `sortFacetValueBy` parameter. - Searching for facet values doesn't work if you have **more than 65 searchable facets and searchable attributes combined**.

    POST /1/indexes/{indexName}/facets/{facetName}/query

  • Read

    Browse for records

    Retrieves records from an index, up to 1,000 per request. Searching returns _hits_ (records augmented with highlighting and ranking details). Browsing returns matching records only. Use browse to export your indices. - The Analytics API doesn't collect data when using `browse`. - Records are ranked by attributes and custom ranking. - There's no ranking for typo tolerance, number of matched words, proximity, or geo distance. Browse requests automatically apply these settings: - `advancedSyntax`: `false` - `attributesToHighlight`: `[]` - `attributesToSnippet`: `[]` - `distinct`: `false` - `enablePersonalization`: `false` - `enableRules`: `false` - `facets`: `[]` - `getRankingInfo`: `false` - `ignorePlurals`: `false` - `optionalFilters`: `[]` - `typoTolerance`: `true` or `false` (`min` and `strict` evaluate to `true`) If you send these parameters with your browse requests, they're ignored.

    POST /1/indexes/{indexName}/browse

  • Read

    Retrieve a record

    Retrieves one record by its object ID. To retrieve more than one record, use the [`objects` operation](https://www.algolia.com/doc/rest-api/search/get-objects).

    GET /1/indexes/{indexName}/{objectID}

  • Read

    Retrieve records

    Retrieves one or more records, potentially from different indices. Records are returned in the same order as the requests.

    POST /1/indexes/*/objects

  • Read

    Retrieve index settings

    Retrieves an object with non-null index settings.

    GET /1/indexes/{indexName}/settings

  • Read

    Retrieve a synonym

    Retrieves a synonym by its ID. To find the object IDs for your synonyms, use the [`search` operation](https://www.algolia.com/doc/rest-api/search/search-synonyms).

    GET /1/indexes/{indexName}/synonyms/{objectID}

  • Read

    Search for synonyms

    Searches for synonyms in your index.

    POST /1/indexes/{indexName}/synonyms/search

  • Read

    List API keys

    Lists all API keys associated with your Algolia application, including their permissions and restrictions.

    GET /1/keys

  • Read

    Retrieve API key permissions

    Gets the permissions and restrictions of an API key. When authenticating with the admin API key, you can request information for any of your application's keys. When authenticating with other API keys, you can only retrieve information for that key, with the description replaced by `<redacted>`.

    GET /1/keys/{key}

  • Read

    Retrieve a rule

    Retrieves a rule by its ID. To find the object ID of rules, use the [`search` operation](https://www.algolia.com/doc/rest-api/search/search-rules).

    GET /1/indexes/{indexName}/rules/{objectID}

  • Read

    Search for rules

    Searches for rules in your index.

    POST /1/indexes/{indexName}/rules/search

  • Read

    Search dictionary entries

    Searches for standard and custom dictionary entries.

    POST /1/dictionaries/{dictionaryName}/search

  • Read

    Retrieve dictionary settings

    Retrieves the languages for which standard dictionary entries are turned off.

    GET /1/dictionaries/*/settings

  • Read

    List available languages

    Lists supported languages with their supported dictionary types and number of custom entries.

    GET /1/dictionaries/*/languages

  • Read

    List user IDs (deprecated)

    Lists the userIDs assigned to a multi-cluster application. Since it can take a few seconds to get the data from the different clusters, the response isn't real-time.

    GET /1/clusters/mapping

  • Read

    Get top user IDs (deprecated)

    Get the IDs of the 10 users with the highest number of records per cluster. Since it can take a few seconds to get the data from the different clusters, the response isn't real-time.

    GET /1/clusters/mapping/top

  • Read

    Retrieve user ID (deprecated)

    Returns the user ID data stored in the mapping. Since it can take a few seconds to get the data from the different clusters, the response isn't real-time.

    GET /1/clusters/mapping/{userID}

  • Read

    List clusters (deprecated)

    Lists the available clusters in a multi-cluster setup.

    GET /1/clusters

  • Read

    Search for user IDs (deprecated)

    Since it can take a few seconds to get the data from the different clusters, the response isn't real-time. To ensure rapid updates, the user IDs index isn't built at the same time as the mapping. Instead, it's built every 12 hours, at the same time as the update of user ID usage. For example, if you add or move a user ID, the search will show an old value until the next time the mapping is rebuilt (every 12 hours).

    POST /1/clusters/mapping/search

  • Read

    Get migration and user mapping status (deprecated)

    To determine when the time-consuming process of creating a large batch of users or migrating users from one cluster to another is complete, this operation retrieves the status of the process.

    GET /1/clusters/mapping/pending

  • Read

    List allowed sources

    Retrieves all allowed IP addresses with access to your application.

    GET /1/security/sources

  • Read

    Retrieve log entries

    The request must be authenticated by an API key with the [`logs` ACL](https://www.algolia.com/doc/guides/security/api-keys/#access-control-list-acl). - Logs are held for the last seven days. - Up to 1,000 API requests per server are logged. - This request counts towards your [operations quota](https://support.algolia.com/hc/articles/17245378392977-How-does-Algolia-count-records-and-operations) but doesn't appear in the logs itself.

    GET /1/logs

  • Read

    Check application task status

    Checks the status of a given application task.

    GET /1/task/{taskID}

  • Read

    Check task status

    Checks the status of a given task. Indexing tasks are asynchronous. When you add, update, or delete records or indices, a task is created on a queue and completed depending on the load on the server. The indexing tasks' responses include a task ID that you can use to check the status.

    GET /1/indexes/{indexName}/task/{taskID}

  • Read

    List indices

    Lists all indices in the current Algolia application. The request follows any index restrictions of the API key you use to make the request.

    GET /1/indexes

  • Write

    Create an API key

    Creates a new API key with specific permissions and restrictions.

    POST /1/keys

  • Write

    Update dictionary settings

    Turns standard stop word dictionary entries on or off for a given language.

    PUT /1/dictionaries/*/settings

  • Write

    Assign or move a user ID (deprecated)

    Assigns or moves a user ID to a cluster. The time it takes to move a user is proportional to the amount of data linked to the user ID.

    POST /1/clusters/mapping

  • Write

    Assign multiple userIDs (deprecated)

    Assigns multiple user IDs to a cluster. **You can't move users with this operation**.

    POST /1/clusters/mapping/batch

  • Write

    Add a source

    Adds a source to the list of allowed sources.

    POST /1/security/sources/append

  • Destructive

    Add a new record (with auto-generated object ID)

    Adds a record to an index or replaces it. - If the record doesn't have an object ID, a new record with an auto-generated object ID is added to your index. - If a record with the specified object ID exists, the existing record is replaced. - If a record with the specified object ID doesn't exist, a new record is added to your index. - If you add a record to an index that doesn't exist yet, a new index is created. To update _some_ attributes of a record, use the [`partial` operation](https://www.algolia.com/doc/rest-api/search/partial-update-object). To add, update, or replace multiple records, use the [`batch` operation](https://www.algolia.com/doc/rest-api/search/batch). This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}

  • Destructive

    Delete an index

    Deletes an index and all its settings. - Deleting an index doesn't delete its analytics data. - If you try to delete a non-existing index, the operation is ignored without warning. - If the index you want to delete has replica indices, the replicas become independent indices. - If the index you want to delete is a replica index, you must first unlink it from its primary index before you can delete it. For more information, see [Delete replica indices](https://www.algolia.com/doc/guides/managing-results/refine-results/sorting/how-to/deleting-replicas).

    DELETE /1/indexes/{indexName}

  • Destructive

    Add or replace a record

    If a record with the specified object ID exists, the existing record is replaced. Otherwise, a new record is added to the index. If you want to use auto-generated object IDs, use the [`saveObject` operation](https://www.algolia.com/doc/rest-api/search/save-object). To update _some_ attributes of an existing record, use the [`partial` operation](https://www.algolia.com/doc/rest-api/search/partial-update-object) instead. To add, update, or replace multiple records, use the [`batch` operation](https://www.algolia.com/doc/rest-api/search/batch).

    PUT /1/indexes/{indexName}/{objectID}

  • Destructive

    Delete a record

    Deletes a record by its object ID. To delete more than one record, use the [`batch` operation](https://www.algolia.com/doc/rest-api/search/batch). To delete records matching a query, use the [`deleteBy` operation](https://www.algolia.com/doc/rest-api/search/delete-by).

    DELETE /1/indexes/{indexName}/{objectID}

  • Destructive

    Delete records matching a filter

    This operation doesn't accept empty filters. This operation is resource-intensive. Use it only if you can't get the object IDs of the records you want to delete. It's more efficient to get a list of object IDs with the [`browse` operation](https://www.algolia.com/doc/rest-api/search/browse), and then delete the records using the [`batch` operation](https://www.algolia.com/doc/rest-api/search/batch). This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/deleteByQuery

  • Destructive

    Delete all records from an index

    Deletes only the records from an index while keeping settings, synonyms, and rules. This operation is resource-intensive and subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/clear

  • Destructive

    Add or update attributes

    Adds new attributes to a record, or updates existing ones. - If a record with the specified object ID doesn't exist, a new record is added to the index **if** `createIfNotExists` is true. - If the index doesn't exist yet, this method creates a new index. - Use first-level attributes only. Nested attributes aren't supported. If you specify a nested attribute, this operation replaces its first-level ancestor. To update attributes without replacing the full record, use these built-in operations. These operations are useful when the initial data isn't available. - `Increment`: increment a numeric attribute. - `Decrement`: decrement a numeric attribute. - `Add`: append a number or string element to an array attribute. - `Remove`: remove all matching number or string elements from an array attribute made of numbers or strings. - `AddUnique`: add a number or string element to an array attribute made of numbers or strings only if it's not already present. - `IncrementFrom`: increment a numeric integer attribute only if the provided value matches the current value. Otherwise, the update is ignored. Example: If you pass an `IncrementFrom` value of 2 for the `version` attribute but the current value is 1, the API ignores the update. If the object doesn't exist, the API only creates it if you pass an `IncrementFrom` value of 0. - `IncrementSet`: increment a numeric integer attribute only if the provided value is greater than the current value. Otherwise, the update is ignored. Example: If you pass an `IncrementSet` value of 2 for the `version` attribute and the current value is 1, the API updates the object. If the object doesn't exist yet, the API only creates it if you pass an `IncrementSet` value greater than 0. Specify an operation by providing an object with the attribute to update as the key and its value as an object with these properties: - `_operation`: the operation to apply on the attribute. - `value`: the right-hand side argument to the operation, for example, increment or decrement step, or a value to add or remove. When updating multiple attributes or using multiple operations targeting the same record, use a single partial update for faster processing. This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/{objectID}/partial

  • Destructive

    Batch indexing operations on one index

    Adds, updates, or deletes records in one index with a single API request. Batching index updates reduces latency and increases data integrity. - Actions are applied in the order they're specified. - Actions are equivalent to the individual API requests of the same name. This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/batch

  • Destructive

    Batch indexing operations on multiple indices

    Adds, updates, or deletes records in multiple indices with a single API request. - Actions are applied in the order they are specified. - Actions are equivalent to the individual API requests of the same name. This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/*/batch

  • Destructive

    Update index settings

    Update the specified index settings. Index settings that you don't specify are left unchanged. Specify `null` to reset a setting to its default value. For best performance, update the index settings before you add new records to your index.

    PUT /1/indexes/{indexName}/settings

  • Destructive

    Create or replace a synonym

    If a synonym with the specified object ID doesn't exist, Algolia adds a new one. Otherwise, the existing synonym is replaced. To add multiple synonyms in a single API request, use the [`batch` operation](https://www.algolia.com/doc/rest-api/search/save-synonyms).

    PUT /1/indexes/{indexName}/synonyms/{objectID}

  • Destructive

    Delete a synonym

    Deletes a synonym by its ID. To find the object IDs of your synonyms, use the [`search` operation](https://www.algolia.com/doc/rest-api/search/search-synonyms).

    DELETE /1/indexes/{indexName}/synonyms/{objectID}

  • Destructive

    Create or replace synonyms

    If a synonym with the `objectID` doesn't exist, Algolia adds a new one. Otherwise, existing synonyms are replaced. This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/synonyms/batch

  • Destructive

    Delete all synonyms

    Deletes all synonyms from the index.

    POST /1/indexes/{indexName}/synonyms/clear

  • Destructive

    Update an API key

    Replaces the permissions of an existing API key. Any unspecified attribute resets that attribute to its default value.

    PUT /1/keys/{key}

  • Destructive

    Delete an API key

    Deletes the API key.

    DELETE /1/keys/{key}

  • Destructive

    Restore an API key

    Restores a deleted API key. Restoring resets the `validity` attribute to `0`. Algolia stores up to 1,000 API keys per application. If you create more, the oldest API keys are deleted and can't be restored.

    POST /1/keys/{key}/restore

  • Destructive

    Create or replace a rule

    If a rule with the specified object ID doesn't exist, it's created. Otherwise, the existing rule is replaced. To create or update more than one rule, use the [`batch` operation](https://www.algolia.com/doc/rest-api/search/save-rules).

    PUT /1/indexes/{indexName}/rules/{objectID}

  • Destructive

    Delete a rule

    Deletes a rule by its ID. To find the object ID for rules, use the [`search` operation](https://www.algolia.com/doc/rest-api/search/search-rules).

    DELETE /1/indexes/{indexName}/rules/{objectID}

  • Destructive

    Create or update rules

    Create or update multiple rules. If a rule with the specified object ID doesn't exist, Algolia creates a new one. Otherwise, existing rules are replaced. This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/rules/batch

  • Destructive

    Delete all rules

    Deletes all rules from the index.

    POST /1/indexes/{indexName}/rules/clear

  • Destructive

    Add or delete dictionary entries

    Adds or deletes multiple entries from your plurals, segmentation, or stop word dictionaries.

    POST /1/dictionaries/{dictionaryName}/batch

  • Destructive

    Delete user ID (deprecated)

    Deletes a user ID and its associated data from the clusters.

    DELETE /1/clusters/mapping/{userID}

  • Destructive

    Replace allowed sources

    Replaces the list of allowed sources.

    PUT /1/security/sources

  • Destructive

    Delete a source

    Deletes a source from the list of allowed sources.

    DELETE /1/security/sources/{source}

  • Destructive

    Copy or move an index

    Copies or moves (renames) an index within the same Algolia application. Notes: - Existing destination indices are overwritten, except for their analytics data. - If the destination index doesn't exist yet, it's created. - This operation is resource-intensive. **Copy** - If the source index doesn't exist, copying creates a new index with 0 records and default settings. - You can't copy the `enableReRanking`, `mode`, and `replicas` settings. - You can't copy to a destination index that already has replicas. - Be aware of the [size limits](https://www.algolia.com/doc/guides/scaling/algolia-service-limits/#application-record-and-index-limits). - For more information, see [Copy indices](https://www.algolia.com/doc/guides/sending-and-managing-data/manage-indices-and-apps/manage-indices/how-to/copy-indices). **Move** - If the source index doesn't exist, moving is ignored without returning an error. - When moving an index, the analytics data keeps its original name, and a new set of analytics data is started for the new name. To access the original analytics in the dashboard, create an index with the original name. - If the destination index has replicas, moving will overwrite the existing index and copy the data to the replica indices. - For more information, see [Move indices](https://www.algolia.com/doc/guides/sending-and-managing-data/manage-indices-and-apps/manage-indices/how-to/move-indices). This operation is subject to [indexing rate limits](https://support.algolia.com/hc/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia).

    POST /1/indexes/{indexName}/operation

search SDK helper

  • Read

    Wait for an API key operation

    Waits for an API key to be added, updated, or deleted.

  • Read

    Wait for operation to complete

    Wait for a task to complete to ensure synchronized index updates. All Algolia write operations are asynchronous. When you make a request for a write operation, for example, to add or update records in your index, Algolia creates a task on a queue and returns a taskID. The task itself runs separately, depending on the server load.

  • Read

    Wait for application-level operation to complete

    Wait for a application-level task to complete.

  • Read

    Get all records from an index

    You can use the browse method to get records from an index—for example, to export your index as a backup. To export all records, use an empty query. Use browse instead of search when exporting records from your index, when ranking, or analytics, isn't important. The Analytics API doesn't collect data when using browse. Don't use this method for building a search UI. Use search instead.

  • Read

    Create secured API keys

    Generates a secured API key without any requests to Algolia's servers. Secured API keys are API keys that you generate on your server without any API request to Algolia. Use secured API keys when you can't update client-side code, such as in mobile apps, or when you need to restrict access to part of an index for each user. When users start searching, your app requests a short-lived secured API key from your server instead of using the Search API key. Your server uses this method to generate the secured API key with restrictions such as filters, index access restrictions, or expiration times, and returns it to your app. The API key gets longer as you add restrictions. You can't create secured API keys from other secured API keys or from your Admin API key. The generated API key can have the same restrictions as the parent API key, or be more restrictive.

  • Read

    Check if an index exists or not

    You can initialize an index with any name. The index is created on Algolia's servers when you add objects or set settings. To prevent accidentally creating new indices, or changing existing indices, you can use the exists method. The exists method returns a boolean that indicates whether an initialized index has been created.

  • Write

    Copies the given `sourceIndexName` records, rules and synonyms to an other Algolia application for the given `destinationIndexName`

    Copies the given `sourceIndexName` records, rules and synonyms to an other Algolia application for the given `destinationIndexName`.

  • Write

    Saves the given array of objects in the given index

    Helper: Saves the given array of objects in the given index. The `chunkedBatch` helper splits this into `batch` requests with at most 1,000 objects each.

  • Write

    Save objects to an Algolia index by leveraging the Transformation pipeline setup using the Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push)

    Helper: Similar to the `saveObjects` method but requires a Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push) to be created first, in order to transform records before indexing them to Algolia. The `region` must have been passed to the client instantiation method.

  • Destructive

    Replace all records in an index

    This method replaces all records in an index without interrupting ongoing searches. It combines [batch](https://www.algolia.com/doc/rest-api/search/batch) and [copy/move](https://www.algolia.com/doc/rest-api/search/operation-index) index operations: 1. Copy settings, synonyms, and rules to a temporary index. 2. Add the records from the `objects` parameter to the temporary index. 3. Replace the original index with the temporary one. Notes: - If there's an error during one of these steps, the temporary index is deleted if your API key has the `deleteIndex` ACL. - If your API key restricts access to specific indices, make sure it also grants access to the temporary index `INDEX_NAME_tmp_*` (replace `INDEX_NAME` with the name of your original index). - This method is subject to [indexing rate limits](https://support.algolia.com/hc/en-us/articles/4406975251089-Is-there-a-rate-limit-for-indexing-on-Algolia). - The response includes the results of the individual API requests. - This method creates a temporary index, which temporarily doubles your record count. - Algolia doesn't count the three days with the highest number of records toward your monthly usage.

  • Destructive

    Replace all records in an index

    Replace all records in your index with a new set of records by using the Transformation pipeline in the Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push). This method replaces all records without downtime. It performs these operations: 1. Copy settings, synonyms, and rules from your original index to a temporary index. 2. Add your new records to the temporary index. 3. Replace your original index with the temporary index. Notes: - Use the `safe` parameter to run these (asynchronous) operations in sequence. - If there's an error during one of these steps, the temporary index isn't deleted. - This operation is rate-limited. - This method creates a temporary index, which temporarily doubles your record count. - Algolia doesn't count the three days with the highest number of records toward your monthly usage. - If you're on a legacy plan (before July 2020), this method counts two operations toward your usage (in addition to the number of records): `copySettings` and `moveIndex`. - The API key you use for this operation must have access to the index `YourIndex` and the temporary index `YourIndex_tmp`.

  • Destructive

    Replace all records in an index

    Helper: Chunks the given `objects` list in subset of 1000 elements max in order to make it fit in `batch` requests.

  • Destructive

    Deletes every records for the given objectIDs

    Helper: Deletes every records for the given objectIDs. The `chunkedBatch` helper splits this into `batch` requests with at most 1,000 objectIDs each.

  • Destructive

    Replaces object content of all the given objects according to their respective `objectID` field

    Helper: Replaces object content of all the given objects according to their respective `objectID` field. The `chunkedBatch` helper splits this into `batch` requests with at most 1,000 objects each.

  • Destructive

    Save objects to an Algolia index by leveraging the Transformation pipeline setup using the Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push)

    Helper: Similar to the `partialUpdateObjects` method but requires a Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push) to be created first, in order to transform records before indexing them to Algolia. The `region` must have been passed to the client instantiation method.

recommend

  • Read

    Retrieve recommendations

    Retrieves recommendations from selected AI models.

    POST /1/indexes/*/recommendations

  • Read

    Retrieve a rule

    Retrieves a Recommend rule that you previously created in the Algolia dashboard.

    GET /1/indexes/{indexName}/{model}/recommend/rules/{objectID}

  • Read

    Check task status

    Checks the status of a given task. Deleting a Recommend rule is asynchronous. When you delete a rule, a task is created on a queue and completed depending on the load on the server. The API response includes a task ID that you can use to check the status.

    GET /1/indexes/{indexName}/{model}/task/{taskID}

  • Read

    Search for rules

    Searches for Recommend rules. Use an empty query to list all rules for this recommendation scenario.

    POST /1/indexes/{indexName}/{model}/recommend/rules/search

  • Destructive

    Delete a rule

    Deletes a Recommend rule from a recommendation scenario.

    DELETE /1/indexes/{indexName}/{model}/recommend/rules/{objectID}

  • Destructive

    Create or update a batch of Recommend Rules

    Create or update a batch of Recommend Rules. Each Recommend Rule is created or updated, depending on whether a Recommend Rule with the same `objectID` already exists. You may also specify `true` for `clearExistingRules`, in which case the batch will atomically replace all the existing Recommend Rules. Recommend Rules are similar to Search Rules, except that the conditions and consequences apply to a [source item](/doc/guides/algolia-recommend/overview/#recommend-models) instead of a query. The main differences are the following: - Conditions `pattern` and `anchoring` are unavailable. - Condition `filters` triggers if the source item matches the specified filters. - Condition `filters` accepts numeric filters. - Consequence `params` only covers filtering parameters. - Consequence `automaticFacetFilters` doesn't require a facet value placeholder (it tries to match the data source item's attributes instead).

    POST /1/indexes/{indexName}/{model}/recommend/rules/batch

analytics

  • Read

    Retrieve top searches

    Returns the most popular searches. For each search, it also includes the average number of hits. If you set the `clickAnalytics` query parameter to `true`, the response also includes: - Tracked searches count Tracked searches are Search API requests with `clickAnalytics` set to `true`. This differs from the response's `count`, which includes searches where `clickAnalytics` is `false`. - Click count - Click-through rate (CTR) - Conversion count - Conversion rate (CR) - Average click position If you set the `revenueAnalytics` query parameter to `true`, the response also includes: - Add-to-cart count - Add-to-cart rate (ATCR) - Purchase count - Purchase rate - Revenue details for each currency **There's a difference between 0% rates and null rates:** - **Null** means there were no queries. Algolia didn't receive any events, so rates are null. - **0% rates** mean there were queries but no [click or conversion events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/searches

  • Read

    Retrieve number of searches

    Retrieves the number of searches within a time range, including a daily breakdown. By default, the analyzed period includes the last eight days including the current day.

    GET /2/searches/count

  • Read

    Retrieve the most frequent searches without results

    Retrieves the 1,000 most frequent searches that produced zero results.

    GET /2/searches/noResults

  • Read

    Retrieve top searches without clicks

    Retrieves the most popular searches that didn't lead to any clicks, from the 1,000 most frequent searches. For each search, it also returns the number of displayed search results that remained unclicked.

    GET /2/searches/noClicks

  • Read

    Retrieve no results rate

    Retrieves the fraction of searches that didn't return any results within a time range, including a daily breakdown. It also returns the count of searches and searches without results used to compute the rates. By default, the analyzed period includes the last eight days including the current day.

    GET /2/searches/noResultRate

  • Read

    Retrieve no click rate

    Retrieves the fraction of searches that didn't lead to any click within a time range, including a daily breakdown. It also returns the number of tracked searches and tracked searches without clicks. By default, the analyzed period includes the last eight days including the current day.

    GET /2/searches/noClickRate

  • Read

    Retrieve top search results

    Retrieves the object IDs of the 1,000 most frequent search results. If you set the `clickAnalytics` query parameter to `true`, the response also includes: - Tracked searches count. Tracked searches are Search API requests with `clickAnalytics` set to `true`. This differs from the response's `count`, which includes searches where `clickAnalytics` is `false`. - Click count - Click-through rate (CTR) - Conversion count - Conversion rate (CR) - Average click position If you set the `revenueAnalytics` parameter to `true`, the response also includes: - Add-to-cart count - Add-to-cart rate (ATCR) - Purchase count - Purchase rate - Revenue details for each currency **There's a difference between 0% rates and null rates:** - **Null** means there were no queries. Algolia didn't receive any events, so rates are null. - **0% rates** mean there were queries but no [click or conversion events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/hits

  • Read

    Retrieve number of users

    Retrieves the number of unique users within a time range, including a daily breakdown. Since it returns the number of unique users, the sum of the daily values might be different from the total number. By default: - Algolia distinguishes search users by their IP address, _unless_ you include a pseudonymous user identifier in your search requests with the `userToken` API parameter or `x-algolia-usertoken` request header. - The analyzed period includes the last eight days including the current day.

    GET /2/users/count

  • Read

    Retrieve top filters

    Retrieves the 1,000 most frequently used filter attributes. These are attributes of your records that you included in the `attributesForFaceting` setting.

    GET /2/filters

  • Read

    Retrieve top filter values

    Retrieves the 1,000 most frequent filter (facet) values for a filter attribute. These are attributes of your records that you included in the `attributesForFaceting` setting.

    GET /2/filters/{attribute}

  • Read

    Retrieve top filters for a search without results

    Retrieves the 1,000 most frequently used filters for a search that didn't return any results. To get the most frequent searches without results, use the [Retrieve searches without results](https://www.algolia.com/doc/rest-api/analytics/get-searches-no-results) operation.

    GET /2/filters/noResults

  • Read

    Retrieve top countries

    Retrieves the countries with the most searches in your index.

    GET /2/countries

  • Read

    Retrieve average click position

    Retrieves the average click position of your search results, including a daily breakdown. The average click position is the average of all clicked search result positions. For example, if users only ever click on the first result for any search, the average click position is 1. By default, the analyzed period includes the last eight days including the current day. An average of `null` when `clickAnalytics` is enabled means Algolia didn't receive any [click events](https://www.algolia.com/doc/guides/sending-events/getting-started) for the queries. The average is `null` until Algolia receives at least one click event.

    GET /2/clicks/averageClickPosition

  • Read

    Retrieve click positions

    Retrieves the positions in the search results and their associated number of clicks. This lets you check how many clicks the first, second, or tenth search results receive. An average of `0` when `clickAnalytics` is enabled means Algolia didn't receive any [click events](https://www.algolia.com/doc/guides/sending-events/getting-started) for the queries.

    GET /2/clicks/positions

  • Read

    Retrieve click-through rate

    Retrieves the click-through rate (CTR) for all your searches with at least one click event, including a daily breakdown. By default, the analyzed period includes the last eight days including the current day. **There's a difference between a 0 and null CTR when `clickAnalytics` is enabled:** - **Null** means there were no queries: since Algolia didn't receive any events, CTR is null. - **0** mean there _were_ queries but no [click events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/clicks/clickThroughRate

  • Read

    Retrieve conversion rate

    Retrieves the conversion rate (CR) for all your searches with at least one conversion event, including a daily breakdown. By default, the analyzed period includes the last eight days including the current day. **There's a difference between a 0 and null CR when `clickAnalytics` is enabled:** - **Null** means there were no queries: since Algolia didn't receive any events, CR is null. - **0** mean there _were_ queries but no [conversion events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/conversions/conversionRate

  • Read

    Retrieve add-to-cart rate

    Retrieves the add-to-cart rate for all your searches with at least one add-to-cart event, including a daily breakdown. By default, the analyzed period includes the last eight days including the current day. The rate is the number of add-to-cart conversion events divided by the number of tracked searches. A search is tracked if it returns a queryID (`clickAnalytics` is `true`). This differs from the response's `count`, which shows the overall number of searches, including those where `clickAnalytics` is `false`. **There's a difference between a 0 and null add-to-cart rate when `clickAnalytics` is enabled:** - **Null** means there were no queries: since Algolia didn't receive any events, the add-to-cart rate is null. - **0** mean there _were_ queries but no [add-to-cart events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/conversions/addToCartRate

  • Read

    Retrieve purchase rate

    Retrieves the purchase rate for all your searches with at least one purchase event, including a daily breakdown. By default, the analyzed period includes the last eight days, including the current day. The rate is purchase conversion events divided by tracked searches. A search is tracked if it returns a query ID (`clickAnalytics` is `true`). This differs from the response's `count`, which includes searches where `clickAnalytics` is `false`. **There's a difference between a 0 and null purchase rate when `clickAnalytics` is enabled:** - **Null** means there were no queries. Algolia didn't receive any events, so the purchase rate is null. - **0** means there were queries but no [purchase conversion events](https://www.algolia.com/doc/guides/sending-events/getting-started) were received.

    GET /2/conversions/purchaseRate

  • Read

    Retrieve revenue data

    Retrieves revenue-related metrics, such as the total revenue or the average order value. To retrieve revenue-related metrics, send purchase events. By default, the analyzed period includes the last eight days including the current day. Revenue is based on purchase conversion events (a conversion event with an `eventSubtype` attribute of `purchase`). The revenue is the `price` attribute multiplied by the `quantity` attribute for each object in the event's `objectData` array.

    GET /2/conversions/revenue

  • Read

    Retrieve update status

    Retrieves the time when the Analytics data for the specified index was last updated. If the index has been recently created or no search has been performed yet the updated time is `null`. The Analytics data is updated every 5&nbsp;minutes.

    GET /2/status

  • Read

    Retrieve the field catalog

    Returns the static catalog of analytics fields, grouped by domain and usage (metrics, filters, groups, distributions). No authentication is required. Use it to discover valid `(domain, kind)` pairs before building the other `/3/patterns/*` queries; two fields are combinable in one query only when their `roots` intersect. Each entry's `requires` lists the ACLs needed when that field is actually used in a query.

    GET /3/patterns/fields

  • Read

    Query a metrics time series

    Returns one time series per `groupBy` combination, each with period `totals` and a per-day metric breakdown. `metrics` and `parameters` are required; `groupBy` and `filters` are optional. Discover valid field kinds per domain with `/3/patterns/fields`.

    POST /3/patterns/timeseries

  • Read

    Query single-row aggregates

    Aggregates the requested `metrics` over the whole period and returns a single object keyed by metric kind. `metrics` and `parameters` are required; `filters` is optional. Discover valid field kinds per domain with `/3/patterns/fields`.

    POST /3/patterns/scalar

  • Read

    Query a grouped table

    Returns `rows`, each a flat object of the requested fields. `metrics` and `parameters` are required; `groupBy`, `filters`, and `orderBy` are optional, though `orderBy` is required when `groupBy` is set. Discover valid field kinds per domain with `/3/patterns/fields`.

    POST /3/patterns/table

  • Read

    Query a numeric distribution

    Buckets one or more numeric fields into histograms and returns an object keyed by `histogram<Field>`, each mapping a bin label to a count. `distributions` and `parameters` are required; `filters` is optional. Discover valid field kinds per domain with `/3/patterns/fields`.

    POST /3/patterns/distribution

insights

  • Write

    Send events

    Sends a list of events to the Insights API. You can include up to 1,000 events in a single request, but the request body must be smaller than 2&nbsp;MB.

    POST /1/events

  • Destructive

    Delete user token events

    Deletes all events related to the specified user token from events metrics and analytics. The deletion is asynchronous, and processed within 48 hours. To delete a personalization user profile, see `Delete a user profile` in the Personalization API.

    DELETE /1/usertokens/{userToken}

abtesting

  • Read

    List all A/B tests (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `GET /3/abtests`. Lists all A/B tests you configured for this application.

    GET /2/abtests

  • Read

    Retrieve A/B test details (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `GET /3/abtests/{id}`. Retrieves the details for an A/B test by its ID.

    GET /2/abtests/{id}

  • Read

    Estimate the sample size and duration of an A/B test (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `POST /3/abtests/estimate`. Given the traffic percentage and the expected effect size, this endpoint estimates the sample size and duration of an A/B test based on historical traffic.

    POST /2/abtests/estimate

  • Write

    Create an A/B test (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `POST /3/abtests`. Creates a new A/B test.

    POST /2/abtests

  • Destructive

    Delete an A/B test (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `DELETE /3/abtests/{id}`. Deletes an A/B test by its ID.

    DELETE /2/abtests/{id}

  • Destructive

    Stop an A/B test (deprecated)

    This endpoint is deprecated. Use the A/B Testing API v3 instead. Replaced by `POST /3/abtests/{id}/stop`. Stops an A/B test by its ID. You can't restart stopped A/B tests.

    POST /2/abtests/{id}/stop

ingestion

  • Read

    List authentication resources

    Retrieves a list of all authentication resources.

    GET /1/authentications

  • Read

    Search for authentication resources

    Searches for authentication resources.

    POST /1/authentications/search

  • Read

    Retrieve an authentication resource

    Retrieves an authentication resource by its ID.

    GET /1/authentications/{authenticationID}

  • Read

    List destinations

    Retrieves a list of destinations.

    GET /1/destinations

  • Read

    Search for destinations

    Searches for destinations.

    POST /1/destinations/search

  • Read

    Retrieve a destination

    Retrieves a destination by its ID.

    GET /1/destinations/{destinationID}

  • Read

    List sources

    Retrieves a list of sources.

    GET /1/sources

  • Read

    Search for sources

    Searches for sources.

    POST /1/sources/search

  • Read

    Retrieve a source

    Retrieve a source by its ID.

    GET /1/sources/{sourceID}

  • Read

    List tasks

    Retrieves a list of tasks.

    GET /2/tasks

  • Read

    Search for tasks

    Searches for tasks.

    POST /2/tasks/search

  • Read

    Retrieve a task

    Retrieves a task by its ID.

    GET /2/tasks/{taskID}

  • Read

    List tasks V1 (deprecated)

    Retrieves a list of tasks using the v1 endpoint. Use `getTasks` instead.

    GET /1/tasks

  • Read

    Search for tasks V1 (deprecated)

    Searches for tasks using the v1 endpoint. Use `searchTasks` instead.

    POST /1/tasks/search

  • Read

    Retrieve a task V1 (deprecated)

    Retrieves a task by its ID using the v1 endpoint. Use `getTask` instead.

    GET /1/tasks/{taskID}

  • Read

    List transformations

    Retrieves a list of transformations.

    GET /1/transformations

  • Read

    Search for transformations

    Searches for transformations.

    POST /1/transformations/search

  • Read

    Retrieve a transformation

    Retrieves a transformation by its ID.

    GET /1/transformations/{transformationID}

  • Read

    List task runs

    Retrieve a list of task runs.

    GET /1/runs

  • Read

    Retrieve a task run

    Retrieve a single task run by its ID.

    GET /1/runs/{runID}

  • Read

    List task run events

    Retrieves a list of events for a task run, identified by its ID.

    GET /1/runs/{runID}/events

  • Read

    Retrieve a task run event

    Retrieves a single task run event by its ID.

    GET /1/runs/{runID}/events/{eventID}

  • Write

    Create an authentication resource

    Creates a new authentication resource.

    POST /1/authentications

  • Write

    Update an authentication resource

    Updates an authentication resource.

    PATCH /1/authentications/{authenticationID}

  • Write

    Create a destination

    Creates a new destination.

    POST /1/destinations

  • Write

    Update a destination

    Updates the destination by its ID.

    PATCH /1/destinations/{destinationID}

  • Write

    Create a source

    Creates a new source.

    POST /1/sources

  • Write

    Validate a source payload

    Validates a source payload to ensure it can be created and that the data source can be reached by Algolia.

    POST /1/sources/validate

  • Write

    Update a source

    Updates a source by its ID.

    PATCH /1/sources/{sourceID}

  • Write

    Validate an update of a source payload

    Validates an update of a source payload to ensure it can be created and that the data source can be reached by Algolia.

    POST /1/sources/{sourceID}/validate

  • Write

    Trigger a stream-listing request

    Triggers a stream-listing request for a source. Triggering stream-listing requests only works with sources with `type: docker` and `imageType: airbyte`.

    POST /1/sources/{sourceID}/discover

  • Write

    Run all tasks linked to a source

    Runs all tasks linked to a source, only available for Shopify, BigCommerce and commercetools sources. Creates one run per task.

    POST /1/sources/{sourceID}/run

  • Write

    Create a task

    Creates a new task.

    POST /2/tasks

  • Write

    Partially update a task

    Partially updates a task by its ID.

    PATCH /2/tasks/{taskID}

  • Write

    Run a task

    Runs a task. You can check the status of task runs with the observability endpoints.

    POST /2/tasks/{taskID}/run

  • Write

    Enable a task

    Enables a task.

    PUT /2/tasks/{taskID}/enable

  • Write

    Disable a task

    Disables a task.

    PUT /2/tasks/{taskID}/disable

  • Write

    Create a task V1 (deprecated)

    Creates a new task using the v1 endpoint. Use `createTask` instead.

    POST /1/tasks

  • Write

    Update a task V1 (deprecated)

    Updates a task by its ID using the v1 endpoint. Use `updateTask` instead.

    PATCH /1/tasks/{taskID}

  • Write

    Run a task V1 (deprecated)

    Runs a task using the v1 endpoint. Use `runTask` instead. You can check the status of task runs with the observability endpoints.

    POST /1/tasks/{taskID}/run

  • Write

    Enable a task V1 (deprecated)

    Enables a task using the v1 endpoint. Use `enableTask` instead.

    PUT /1/tasks/{taskID}/enable

  • Write

    Disable a task V1 (deprecated)

    Disables a task using the v1 endpoint. Use `disableTask` instead.

    PUT /1/tasks/{taskID}/disable

  • Write

    Create a transformation

    Creates a new transformation.

    POST /1/transformations

  • Write

    Try a transformation before creating it

    Try a transformation before creating it.

    POST /1/transformations/try

  • Write

    Update a transformation

    Updates a transformation by its ID.

    PUT /1/transformations/{transformationID}

  • Write

    Try a transformation before updating it

    Try a transformation before updating it.

    POST /1/transformations/{transformationID}/try

  • Destructive

    Push records by indexName

    Pushes records through the Pipeline, directly to an index. You can make the call synchronous by providing the `watch` parameter, for asynchronous calls, you can use the observability endpoints and/or debugger dashboard to see the status of your task. If you want to leverage the [pre-indexing data transformation](https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/how-to/transform-your-data), this is the recommended way of ingesting your records. This method is similar to `pushTask`, but requires an `indexName` instead of a `taskID`. If zero or many tasks are found, an error will be returned.

    POST /1/push/{indexName}

  • Destructive

    Delete an authentication resource

    Deletes an authentication resource. You can't delete authentication resources that are used by a source or a destination.

    DELETE /1/authentications/{authenticationID}

  • Destructive

    Delete a destination

    Deletes a destination by its ID. You can't delete destinations that are referenced in tasks.

    DELETE /1/destinations/{destinationID}

  • Destructive

    Delete a source

    Deletes a source by its ID. You can't delete sources that are referenced in tasks.

    DELETE /1/sources/{sourceID}

  • Destructive

    Fully update a task

    Fully updates a task by its ID, use partialUpdateTask if you only want to update a subset of fields.

    PUT /2/tasks/{taskID}

  • Destructive

    Delete a task

    Deletes a task by its ID.

    DELETE /2/tasks/{taskID}

  • Destructive

    Push records by task ID

    Pushes records through the pipeline, directly to an index. You can make the call synchronous by providing the `watch` parameter, for asynchronous calls, you can use the observability endpoints or the debugger dashboard to see the status of your task. If you want to transform your data before indexing, this is the recommended way of ingesting your records. This method is similar to `push`, but requires a `taskID` instead of a `indexName`, which is useful when many `destinations` target the same `indexName`.

    POST /2/tasks/{taskID}/push

  • Destructive

    Delete a task (deprecated)

    Deletes a task by its ID using the v1 endpoint. Use `deleteTask` instead.

    DELETE /1/tasks/{taskID}

  • Destructive

    Delete a transformation

    Deletes a transformation by its ID.

    DELETE /1/transformations/{transformationID}

ingestion SDK helper

  • Destructive

    Replace all records in an index

    Helper: Chunks the given `objects` list in subset of 1000 elements max in order to make it fit in `push` requests by leveraging the Transformation pipeline setup in the Push connector (https://www.algolia.com/doc/guides/sending-and-managing-data/send-and-update-your-data/connectors/push).

monitoring

  • Read

    Retrieve status of all clusters

    Retrieves the status of all Algolia clusters and instances.

    GET /1/status

  • Read

    Retrieve cluster status

    Retrieves the status of selected clusters.

    GET /1/status/{clusters}

  • Read

    Retrieve all incidents

    Retrieves known incidents for all clusters.

    GET /1/incidents

  • Read

    Retrieve cluster incidents

    Retrieves known incidents for the selected clusters.

    GET /1/incidents/{clusters}

  • Read

    Retrieve servers

    Retrieves the servers that belong to clusters. The response depends on whether you authenticate your API request: - With authentication, the response lists the servers assigned to your Algolia application's cluster. - Without authentication, the response lists the servers for all Algolia clusters.

    GET /1/inventory/servers

  • Read

    Retrieve search latency times

    Retrieves the average latency for search requests for selected clusters.

    GET /1/latency/{clusters}

  • Read

    Retrieve indexing monitoring latency

    Retrieves indexing latency metrics for selected clusters. This endpoint is intended for infrastructure-level monitoring and availability checks. The returned value reflects latency measured on Algolia's internal monitoring index and is reported in milliseconds. This metric isn't intended to represent the indexing performance of an individual application or index. To measure when an indexing operation has completed for your application, use the `waitTask` method.

    GET /1/indexing/{clusters}

  • Read

    Test the reachability of clusters

    Test whether clusters are reachable or not.

    GET /1/reachability/{clusters}/probes

  • Read

    Retrieve metrics

    Retrieves metrics related to your Algolia infrastructure, aggregated over a selected time window. Access to this API is available as part of the [Premium or Elevate plans](https://www.algolia.com/pricing). You must authenticate requests with the `x-algolia-application-id` and `x-algolia-api-key` headers (using the Monitoring API key).

    GET /1/infrastructure/{metric}/period/{period}

personalization

  • Read

    Retrieve a user profile

    Retrieves a user profile and their affinities for different facets.

    GET /1/profiles/personalization/{userToken}

  • Read

    Retrieve the personalization strategy

    Retrieves the current personalization strategy.

    GET /1/strategies/personalization

  • Write

    Define the personalization strategy

    Creates a new personalization strategy.

    POST /1/strategies/personalization

  • Destructive

    Delete a user profile

    Deletes a user profile. The response includes a date and time when the user profile can safely be considered deleted.

    DELETE /1/profiles/{userToken}

query-suggestions

  • Read

    List configurations

    Retrieves all Query Suggestions configurations of your Algolia application.

    GET /1/configs

  • Read

    Retrieve a configuration

    Retrieves a single Query Suggestions configuration by its index name.

    GET /1/configs/{indexName}

  • Read

    Retrieve configuration status

    Reports the status of a Query Suggestions index.

    GET /1/configs/{indexName}/status

  • Read

    Retrieve logs

    Retrieves the logs for a single Query Suggestions index.

    GET /1/logs/{indexName}

  • Write

    Create a configuration

    Creates a new Query Suggestions configuration. You can have up to 100 configurations per Algolia application.

    POST /1/configs

  • Write

    Update a configuration

    Updates a QuerySuggestions configuration.

    PUT /1/configs/{indexName}

  • Destructive

    Delete a configuration

    Deletes a Query Suggestions configuration. Deleting only removes the configuration and stops updates to the Query Suggestions index. To delete the Query Suggestions index itself, use the Search API and the `Delete an index` operation.

    DELETE /1/configs/{indexName}

composition SDK helper

  • Read

    Send requests to the Algolia REST API

    This method lets you send requests to the Algolia REST API.

  • Read

    Switch the API key used to authenticate requests

    Switch the API key used to authenticate requests.

  • Read

    Wait for operation to complete

    Wait for a task to complete to ensure synchronized composition updates. All Algolia write operations are asynchronous. When you make a request for a write operation, for example, to upsert or delete a composition, Algolia creates a task on a queue and returns a taskID. The task itself runs separately, depending on the server load.

  • Write

    Send requests to the Algolia REST API

    This method lets you send requests to the Algolia REST API.

  • Write

    Send requests to the Algolia REST API

    This method lets you send requests to the Algolia REST API.

  • Destructive

    Send requests to the Algolia REST API

    This method lets you send requests to the Algolia REST API.

composition

  • Read

    Search for facet values

    Searches for values of a specified facet attribute on the composition's main source's index. - By default, facet values are sorted by decreasing count. You can adjust this with the `sortFacetValueBy` parameter. - Searching for facet values doesn't work if you have **more than 65 searchable facets and searchable attributes combined**.

    POST /1/compositions/{compositionID}/facets/{facetName}/query

  • Read

    List compositions

    Lists all compositions in the current Algolia application.

    GET /1/compositions

  • Read

    Retrieve a composition

    Retrieve a single composition in the current Algolia application.

    GET /1/compositions/{compositionID}

  • Read

    Retrieve a rule

    Retrieves a rule by its ID. To find the object ID of a rule, use the [`search` operation](https://www.algolia.com/doc/rest-api/composition/search-composition-rules).

    GET /1/compositions/{compositionID}/rules/{objectID}

  • Read

    Search for composition rules

    Searches for composition rules in your index.

    POST /1/compositions/{compositionID}/rules/search

  • Read

    Check task status

    Checks the status of a given task.

    GET /1/compositions/{compositionID}/task/{taskID}

  • Destructive

    Update and insert (upsert) a composition

    Update and insert a composition in the current Algolia application.

    PUT /1/compositions/{compositionID}

  • Destructive

    Delete a composition

    Delete a composition from the current Algolia application.

    DELETE /1/compositions/{compositionID}

  • Destructive

    Batch action to multiple compositions

    Adds, updates, or deletes compositions with a single API request.

    POST /1/compositions/*/batch

  • Destructive

    Set or update the "sortingStrategy" configuration for an existing composition

    Updates the "sortingStrategy" field of an existing composition. This endpoint lets you create a new sorting strategy mapping or replace the configured one. The provided sorting indices must be associated indices or replicas of the main targeted index. This endpoint can't validate whether the sort index is related to the composition's main index. Validation fails at runtime if the index you updated isn't related. The update is applied to the specified composition within the current Algolia application and returns a taskID that can be used to track the operation’s completion.

    POST /1/compositions/{compositionID}/sortingStrategy

  • Destructive

    Add or update a composition rule

    If a composition rule with the provided ID already exists, it's replaced. Otherwise, a new one is added.

    PUT /1/compositions/{compositionID}/rules/{objectID}

  • Destructive

    Delete a Composition Rule

    Delete a Composition Rule from the specified Composition ID.

    DELETE /1/compositions/{compositionID}/rules/{objectID}

  • Destructive

    Create or update or delete composition rules

    Create or update or delete multiple composition rules.

    POST /1/compositions/{compositionID}/rules/batch

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

Reference date: October 29, 2026. Algolia API docs.