Integration

Peaka integration

Peaka is a zero-ETL data integration platform that exposes connected sources (SaaS apps, databases, files) and its own tables as one SQL-queryable project. Run SQL or saved queries and read rows, browse catalogs, schemas, tables and columns, manage saved and materialized queries, create Peaka Tables and BI Tables and their columns, cache source tables and trigger or monitor their syncs, and export a query or table to CSV or JSONL download links. Connections and their credentials, API keys, catalog creation and sharing, query folders, semantic catalogs, metadata, the AI agent, embedded UI sessions, CSV import and the deprecated project routes are not exposed, nor workspace and project updates or deletion. This integration calls the US zone (partner.peaka.studio). Authentication is a Peaka API key sent as Authorization: Bearer <key>: a Project API Key reaches the project-scoped tools, a Partner API Key also reaches organizations, workspaces and projects.

Connect Peaka How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Peaka.Built from the 55 Peaka tools below, on the connection you add once.

What you can do with Peaka

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.

Cache, 6 operations

create_cacheCreate cachePOSTCaches a source table in Peaka for faster queries and starts its first sync at once. Without schedules it syncs once only. Check get_table_cache_info first.
delete_cacheDelete cacheDELETEPermanently deletes a cache: cancels its running and scheduled syncs, drops the cached data and removes its settings and history. Queries then read the source live.
get_cache_settingsGet cache settingsGETReturns a cache's settings: the cached catalog, schema and table and its incremental and full refresh schedules.
start_cache_full_refreshStart cache full refreshPOSTStarts a full refresh of a cache now: drops the cached data and re-reads the whole source table (costlier than an incremental sync). Follow it with get_cache_status.
start_cache_incremental_syncStart cache incremental syncPOSTStarts an incremental sync of a cache now (only rows inserted, updated or deleted since the last successful sync). Follow it with get_cache_status.
update_cache_settingsUpdate cache settingsPUTChanges a cache's sync schedules (the cached table cannot change); a changed schedule replaces the running one, type NONE disables that mode. Returns the cache.

Query, 6 operations

create_queryCreate queryPOSTSaves a SQL query in the project (PLAIN, or MATERIALIZED: computed at once and queryable as peaka.mtquery.<name>) and returns it with its id.
delete_queryDelete queryDELETEPermanently deletes a saved query.
execute_queryExecute queryPOSTRuns SQL (statement) or a saved query (id or qualifiedName) or a query-builder request (from) and returns columns and rows in data. Send exactly one of the four.
get_queryGet queryGETReturns a saved query: id, display name, SQL name, SQL, type (PLAIN or MATERIALIZED), schedule and folder path.
list_queriesList queriesGETLists the saved queries of a project with id, name, SQL, type (PLAIN or MATERIALIZED), schedule and folder path. Run one with execute_query.
update_queryUpdate queryPUTUpdates a saved query's display name, SQL or schedule; fields left out keep their current value. Returns the query.

Export, 5 operations

cancel_exportCancel exportDELETECancels a PENDING or RUNNING export (stops its query and removes partial files). Does nothing on a finished job.
export_queryExport queryPOSTExports a saved query to CSV or JSONL files and returns their short-lived download links in files. Read status: SUCCEEDED, or FAILED, CANCELLED. Waits for the job and returns the finished result; gives up after 10 minutes.
export_tableExport tablePOSTExports a table to CSV or JSONL files and returns their short-lived download links in files. Read status: SUCCEEDED, or FAILED, CANCELLED. Waits for the job and returns the finished result; gives up after 10 minutes.
get_exportGet exportGETReturns an export job's status and, for a SUCCEEDED job you own, fresh short-lived download links in files (others get downloadable false).
list_exportsList exportsGETLists the project's export jobs, newest first, with status, format, row count and source. Download links are left out: get_export returns them.

Bi table column, 4 operations

add_bi_table_columnsAdd bi table columnsPOSTAdds one or more columns to a BI Table and returns the resulting column definitions.
delete_bi_table_columnDelete bi table columnDELETEPermanently removes a column and all its stored data from a BI Table. Irreversible.
list_bi_table_columnsList bi table columnsGETLists the columns of a BI Table: name, data type, display name, default value and constraints.
update_bi_table_columnUpdate bi table columnPUTUpdates a column of a BI Table (name, display name, data type, default value, constraints) and returns the table's columns. Send the full column definition.

Peaka table column, 4 operations

add_peaka_table_columnsAdd peaka table columnsPOSTAdds one or more columns to a Peaka Table and returns the resulting column definitions.
delete_peaka_table_columnDelete peaka table columnDELETEPermanently removes a column and all its stored data from a Peaka Table. Irreversible.
list_peaka_table_columnsList peaka table columnsGETLists the columns of a Peaka Table: name, data type, display name, default value and constraints.
update_peaka_table_columnUpdate peaka table columnPUTUpdates a column of a Peaka Table (name, display name, data type, default value, constraints) and returns the table's columns. Send the full column definition.

Bi table, 3 operations

create_bi_tableCreate bi tablePOSTCreates an empty BI Table named tableName in the project and returns it; add columns with add_bi_table_columns.
delete_bi_tableDelete bi tableDELETEPermanently deletes a BI Table and all its rows, columns and schema definitions. Irreversible.
list_bi_tablesList bi tablesGETLists the BI Tables (column-oriented tables for large datasets of over one million records) of the project, with catalog, schema and table names.

Peaka table, 3 operations

create_peaka_tableCreate peaka tablePOSTCreates an empty Peaka Table named tableName in the project and returns it; add columns with add_peaka_table_columns.
delete_peaka_tableDelete peaka tableDELETEPermanently deletes a Peaka Table and all its rows, columns and schema definitions. Irreversible.
list_peaka_tablesList peaka tablesGETLists the Peaka Tables (relational tables stored in Peaka, catalog peaka, schema table) of the project, with catalog, schema and table names.

Project, 3 operations

create_projectCreate projectPOSTCreates a project in a workspace and returns it with its id (the projectId of the data tools). Needs a Partner API Key.
get_projectGet projectGETReturns a project: id, name, description, domain, owner, workspace and creation time. Needs a Partner API Key.
list_projectsList projectsGETLists the projects of a workspace. Pass id as projectId to every data tool (catalogs, queries, tables, caches, exports). Needs a Partner API Key.

Workspace, 3 operations

create_workspaceCreate workspacePOSTCreates a workspace in an organization and returns it with its id. Needs a Partner API Key.
get_workspaceGet workspaceGETReturns a workspace: id, name, organizationId, description, creation time and whether it is the default one. Needs a Partner API Key.
list_workspacesList workspacesGETLists the workspaces of an organization (a workspace groups projects). Pass id as workspaceId to list_projects. Needs a Partner API Key.

Cache status, 2 operations

get_cache_statusGet cache statusGETReturns a cache's status (NOT_INITIALIZED, RUNNING, COMPLETED, FAILED or CANCELLED) with its last incremental sync and last full refresh. Set excludeLogs while polling.
list_cache_statusesList cache statusesGETLists the status of every cache of the project, with the last incremental sync and last full refresh of each (status, record counts, errors).

Catalog, 2 operations

get_catalogGet catalogGETReturns a catalog: id, name, display name, catalog type and connection id.
list_catalogsList catalogsGETLists the catalogs of a project: one per connected data source, plus the built-in internal (Peaka Tables and BI Tables) and query (saved queries) catalogs. Pass id to list_schemas.

Materialized query, 2 operations

cancel_materialized_query_refreshCancel materialized query refreshPOSTCancels the refresh of a materialized query that is in progress, if any. Returns nothing.
refresh_materialized_queryRefresh materialized queryPOSTStarts an immediate refresh of a materialized query (its scheduled workflow when it has a schedule, else a one-time refresh). Returns nothing; follow it with get_materialized_query_status.

Materialized query status, 2 operations

get_materialized_query_statusGet materialized query statusGETReturns the refresh status and schedule of one materialized query: status of the most recent refresh, last and next refresh times. Use it after refresh_materialized_query.
list_materialized_query_statusesList materialized query statusesGETLists the refresh status and schedule of every materialized query of the project: last and next refresh times and the status of the most recent refresh.

Organization, 2 operations

get_organizationGet organizationGETReturns an organization: id, name, owner, description and creation time. Needs a Partner API Key.
list_organizationsList organizationsGETLists the organizations the Partner API Key can reach, with id and name. Pass id as organizationId to list_workspaces. Needs a Partner API Key.

Cache execution, 1 operation

list_cache_executionsList cache executionsGETLists the past sync executions of a cache, newest first, with status, record counts, errors and timestamps. Page with limit and offset.

Cache full refresh, 1 operation

cancel_cache_full_refreshCancel cache full refreshPOSTSends a cancellation to the running full refresh of a cache (best-effort). Returns nothing; check get_cache_status to confirm.

Cache incremental sync, 1 operation

cancel_cache_incremental_syncCancel cache incremental syncPOSTSends a cancellation to the running incremental sync of a cache (best-effort). Returns nothing; check get_cache_status to confirm.

Catalog object, 1 operation

search_projectSearch projectPOSTSearches the catalogs, schemas and tables of a project whose name contains query (case-insensitive), optionally within one catalog or schema. Read meta.page.nextOffset to page.

Column, 1 operation

list_catalog_table_columnsList catalog table columnsGETLists the columns of a catalog table: name, data type, display name, default value and constraints. Columns prefixed with _q_ are virtual query parameters passed to the source.

Schema, 1 operation

list_schemasList schemasGETLists the schemas of a catalog (logical groups of tables; for external connectors usually a database or namespace of the source). Pass schemaName to list_catalog_tables.

Table, 1 operation

list_catalog_tablesList catalog tablesGETLists the tables of a catalog schema with their caching metadata (isCacheable, supportedCacheTypes, isCached) and whether they need _q_ query parameters (isDynamicTable).

Table cache, 1 operation

get_table_cache_infoGet table cache infoGETTells whether a table has an active cache (isCached), whether it can be cached (isCacheable) and with which cache types (supportedCacheTypes), before create_cache.

Use Peaka from Claude Code, Cursor or Codex

Connect Peaka in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Peaka 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 Peaka

Create your workspace
Sign up free. Nothing to install.
Connect Peaka once
Add your Peaka 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.