Integration

OpenSearch integration

OpenSearch is a community-driven, open-source search and analytics suite. It provides distributed search, aggregations, and analytics capabilities for log analysis, real-time application monitoring, and clickstream analytics. Features include full-text search, structured queries, aggregations, alerting, anomaly detection, and vector search for AI applications. Compatible with Elasticsearch clients and APIs.

Connect OpenSearch How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in OpenSearch.Built from the 25 OpenSearch tools below, on the connection you add once.

Common OpenSearch use cases

disaster recovery
data archival
regular backups
bulk data import
full-text search
adding new fields
capacity planning
dashboard metrics
dashboard queries
bulk field updates

What you can do with OpenSearch

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Document, 11 operations

bulk_operationsBulk operationsPOSTExecutes multiple index, create, update, and delete operations in a single request. Each operation requires an action line and optional data line in NDJSON format. Significantly more efficient than individual API calls for batch processing.
count_documentsCount documentsPOSTReturns the count of documents matching a query without retrieving the documents. More efficient than search when only the count is needed. Supports full Query DSL.
delete_by_queryDelete by queryPOSTDeletes all documents matching a query. Processes documents in batches using scroll. Returns count of deleted documents and any failures. Set wait_for_completion=false for async execution.
delete_documentDelete documentDELETEDeletes a document from an index by its ID. The document is permanently removed. Returns the deletion result including version information and whether the document was found.
get_documentGet documentGETRetrieves a document by its ID from the specified index. Returns the document source (_source) along with metadata including version and index information. Use _source parameter to limit returned fields.
index_documentIndex documentPOSTIndexes a JSON document into the specified index. Creates the document if it doesn't exist or replaces it if it does (when ID is provided). Returns the document ID, version, and index result status.
multi_searchMulti searchPOSTExecutes multiple search requests in a single API call using NDJSON format. Each search has a header line (index, preferences) and body line (query). Returns array of responses corresponding to each search request.
scroll_searchScroll searchPOSTContinues scrolling through search results using a scroll context. First search with scroll parameter returns scroll_id. Subsequent calls with scroll_id retrieve next batch of results. Use for processing large result sets.
searchSearchPOSTExecutes a search query against one or more indices. Supports Query DSL for full-text search, term queries, range filters, boolean logic, and nested queries. Returns matching documents with scores, total hits, and optional aggregations.
update_by_queryUpdate by queryPOSTUpdates all documents matching a query using a script. Processes in batches with scroll. Script can modify, add, or remove fields. Returns count of updated documents and failures.
update_documentUpdate documentPOSTUpdates an existing document with partial changes or using a script. The doc field contains fields to merge with the existing document. Scripts enable complex transformations like incrementing counters.

Index, 3 operations

create_indexCreate indexPUTCreates a new OpenSearch index with specified settings, mappings, and aliases. Settings control shard configuration, replicas, and analyzers. Mappings define field types and indexing behavior. Returns acknowledgment with index name on success.
delete_indexDelete indexDELETEPermanently deletes one or more indices including all documents, mappings, and settings. Supports wildcards and comma-separated lists. This operation cannot be undone - all data is permanently lost.
get_indexGet indexGETRetrieves detailed information about one or more indices including settings, mappings, and aliases. Supports wildcards and comma-separated lists. Use for inspecting index configuration and field definitions.

Snapshot, 3 operations

create_snapshotCreate snapshotPUTCreates a snapshot of specified indices in a repository. Snapshots are incremental - only changes since the last snapshot are stored. Can include global state for complete cluster backup.
get_snapshotGet snapshotGETRetrieves information about snapshots including state, start/end times, indices included, and shard statistics. Use _all to list all snapshots in a repository.
restore_snapshotRestore snapshotPOSTRestores indices from a snapshot to the cluster. Can restore specific indices, rename during restore, and optionally include aliases and global state. Cannot restore to existing open indices with same name.

Alias, 2 operations

create_aliasCreate aliasPOSTManages index aliases with add, remove, and remove_index actions atomically. Aliases provide alternative names for indices, enable zero-downtime reindexing, and can include filters for data partitioning.
get_aliasesGet aliasesGETRetrieves alias definitions showing which indices each alias points to, along with filters and routing settings. Use for understanding alias configuration and debugging.

Cluster, 2 operations

cluster_healthCluster healthGETReturns cluster health status: green (all shards allocated), yellow (primary shards allocated, replicas missing), or red (primary shards missing). Includes node counts, shard statistics, and pending tasks.
cluster_statsCluster statsGETReturns comprehensive cluster statistics including node counts, index counts, document counts, storage size, and JVM/OS metrics. Aggregates data across all nodes in the cluster.

Mapping, 2 operations

get_mappingGet mappingGETRetrieves mapping definitions for one or more indices showing all field types, analyzers, and properties. Use for understanding index structure and debugging field type issues.
put_mappingPut mappingPUTAdds new fields to an existing index mapping or updates field properties where allowed. Cannot change existing field types - requires reindexing. Returns acknowledgment on success.

Index setting, 1 operation

update_index_settingsUpdate index settingsPUTUpdates dynamic settings for indices. Can change number_of_replicas, refresh_interval, and other dynamic settings. Static settings like number_of_shards cannot be changed after creation.

Snapshot repository, 1 operation

register_snapshot_repositoryRegister snapshot repositoryPUTRegisters a snapshot repository for backup and restore operations. Supports filesystem (fs), Amazon S3 (s3), HDFS (hdfs), and Azure storage backends. Repository must be registered before creating snapshots.

Use OpenSearch from Claude Code, Cursor or Codex

Connect OpenSearch in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its OpenSearch operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect OpenSearch

Create your workspace
Sign up free. Nothing to install.
Connect OpenSearch once
Add your OpenSearch connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

OpenSearch integration FAQ

What can you do with the OpenSearch integration?

LiveContext exposes 25 operations on OpenSearch, covering document, index, snapshot and alias. Common uses include disaster recovery, data archival and regular backups.