Algolia

Agent-callable tools for the Algolia Search and Recommend REST APIs. Index records into a search index, search and browse them, retrieve records by object ID, and manage the configuration that shapes relevance — index settings, synonyms, and query rules — plus read AI recommendations. Covers the full record lifecycle (write / read / search / delete, single and batch), index and index-config management, and the Recommend read surface.

Independent, unofficial connector for Algolia. Not affiliated with, endorsed by, or sponsored by Algolia. "Algolia" is a trademark of its owner, used only to identify the service this connector works with.

When to use this

• Search and retrieve — full-text search over an index (with filters and facets), federated multi-index search, cursor browse for export, get records by object ID, and facet-value search.
• Index records — add, replace, partial-update, and delete records, individually or in batches, across one or many indices.
• Manage indices and relevance config — list/delete/copy/move indices, read and update settings, and CRUD synonyms and query rules.
• Recommendations — read AI recommendations (related products, frequently-bought-together, trending, looking-similar) and the Recommend rules that curate them.

Writes are asynchronous — each returns a taskID; use getTask to confirm a write is applied before asserting it landed.

Setup

This is an agentskills.io skill.

If the connector has not been installed as a skill yet, install it first with npx skills add zapier/connectors --skill algolia (or your harness's own skill-install mechanism), then continue here. Installing the skill copies these files, not dependencies. Before running the CLI, a local MCP server, or zapier-sdk auth commands, run npm install --omit=dev here once. Importing the published package as a dependency in your own project instead? That npm install already resolves everything — see references/use-as-sdk.md.

Want the actual repo source instead — to browse references/, run this connector's tests, or hack on it? See README.md for a scoped git clone.

The connector runs on Node.js 22.18+. Pick the reference that matches how you're running it, and load it before doing anything else:

You have...	Load
An MCP-aware client — tools may already be loaded (e.g. mcp__algolia__<tool>), or you can register a local server yourself (or guide the user to)	references/use-as-mcp.md
Terminal / subprocess access (you can run node)	references/use-as-cli.md
Only your own code, importing this package as a dependency	references/use-as-sdk.md
No tool access, no terminal, no ability to import this package — you write your own code that calls the Algolia API directly (e.g. a code-execution sandbox)	references/use-as-recipe.md

Scripts

All scripts share the single algolia connection. Almost every tool takes an indexName — get it from listIndices. Object/synonym/rule IDs come from the corresponding search/browse tool.

Script	Script name	Connections	Description
scripts/searchIndex.ts	searchIndex	algolia	Search one index for records matching a query, with filters and facets.
scripts/searchMultipleIndices.ts	searchMultipleIndices	algolia	Run several search queries in one request (federated / multi-index).
scripts/browseObjects.ts	browseObjects	algolia	Cursor-traverse every record in an index (export / reindex).
scripts/getObject.ts	getObject	algolia	Retrieve a single record by object ID.
scripts/getObjects.ts	getObjects	algolia	Retrieve multiple records by object ID, optionally across indices.
scripts/searchForFacetValues.ts	searchForFacetValues	algolia	Search the values of one faceted attribute.
scripts/saveObject.ts	saveObject	algolia	Add a record; Algolia assigns an object ID if omitted.
scripts/addOrUpdateObject.ts	addOrUpdateObject	algolia	Add or fully replace a record at a specific object ID (upsert).
scripts/partialUpdateObject.ts	partialUpdateObject	algolia	Update specific attributes on a record without replacing it.
scripts/deleteObject.ts	deleteObject	algolia	Delete a single record by object ID.
scripts/batch.ts	batch	algolia	Run many record operations against one index in one request.
scripts/multipleBatch.ts	multipleBatch	algolia	Run record operations across several indices in one request.
scripts/deleteBy.ts	deleteBy	algolia	Delete records in an index matching a filter (a filter is required).
scripts/clearObjects.ts	clearObjects	algolia	Delete all records in an index, keeping its config.
scripts/listIndices.ts	listIndices	algolia	List the indices in the application.
scripts/deleteIndex.ts	deleteIndex	algolia	Permanently delete an index (records + config).
scripts/copyOrMoveIndex.ts	copyOrMoveIndex	algolia	Copy or move (rename) an index; overwrites the destination.
scripts/getSettings.ts	getSettings	algolia	Read an index's settings (relevance, faceting, ranking).
scripts/setSettings.ts	setSettings	algolia	Update an index's settings (merged with existing).
scripts/saveSynonym.ts	saveSynonym	algolia	Create or replace a single synonym.
scripts/getSynonym.ts	getSynonym	algolia	Fetch a single synonym by ID.
scripts/searchSynonyms.ts	searchSynonyms	algolia	Search or list synonyms in an index.
scripts/deleteSynonym.ts	deleteSynonym	algolia	Delete a single synonym by ID.
scripts/clearSynonyms.ts	clearSynonyms	algolia	Delete all synonyms in an index.
scripts/saveRule.ts	saveRule	algolia	Create or replace a single query rule.
scripts/getRule.ts	getRule	algolia	Fetch a single query rule by ID.
scripts/searchRules.ts	searchRules	algolia	Search or list query rules in an index.
scripts/deleteRule.ts	deleteRule	algolia	Delete a single query rule by ID.
scripts/getRecommendations.ts	getRecommendations	algolia	Read AI recommendations from a Recommend model.
scripts/getRecommendRule.ts	getRecommendRule	algolia	Fetch a single Recommend rule by ID.
scripts/searchRecommendRules.ts	searchRecommendRules	algolia	Search or list Recommend rules for an index + model.
scripts/getTask.ts	getTask	algolia	Check an async indexing task's status (poll until published).

Disambiguation & refusals

• Resolve names to IDs before writing. Most write tools (addOrUpdateObject, partialUpdateObject, deleteObject, saveSynonym/deleteSynonym, saveRule/deleteRule) act on an objectID or an indexName, not a human name. Resolve first: use listIndices to pick the index, and searchIndex / browseObjects / searchSynonyms / searchRules to find the exact ID. If a search returns exactly one clear match, act on it; if two or more plausibly match, stop and ask which — list the candidates with a distinguishing field (e.g. a title attribute or the rule/synonym body). Never guess an ID or silently pick among ties.
• Don't fabricate unsupported operations. This connector does not manage API keys, edit language dictionaries, write Recommend rules (getRecommendRule/searchRecommendRules are read-only), or run Insights/Analytics/Personalization. If asked for one of these, say it's unsupported — don't substitute another tool and report success for something you didn't do.
• Destructive tools are irreversible. deleteBy (requires a filter — an empty filter is refused; use clearObjects to intentionally empty an index), clearObjects, deleteIndex, and copyOrMoveIndex (overwrites the destination) cannot be undone. Confirm the target index/filter before calling.

Auth

Every shape passes auth as one connection selector, not the secret — a [<resolver>:]<value> string. Every connector accepts zapier:<connection-id> (Zapier-managed auth — routes through Zapier's auth, retries, and governance layer); some also accept one or more direct-token resolvers (naming and count vary per connector) — check this connector's own resolvers rather than assuming. The <resolver>: prefix is optional; a bare value goes to the first resolver that claims it — a UUID-shaped bare value always claims zapier:. Each script declares the connections it needs and the resolvers each accepts. The exact syntax for passing a connection (and how to see this connector's resolver list) differs by shape — see the reference you loaded above.

Checking what's already configured first? Don't dump environment values to do it — env or env | grep <name> prints the value along with the name, leaking a live credential into the transcript if one is set. Check names only (env | cut -d= -f1 | grep -i <name>) or test a known name directly ([ -n "$VAR_NAME" ]).

No connection yet? Pick one — and follow the reference's own flow to obtain it; never just ask the user for a connection id or token as if they already have one memorized:

	Load
Pass the credential directly	references/use-without-zapier.md
Route it through a Zapier connection	references/use-with-zapier.md

Output format

Every script returns a { data, meta } envelope:

• data — the script's result (the shape its outputSchema declares; see the reference you loaded above for how to inspect a script's exact schema in your shape).
• meta.outputDataValidation — what validating data did:
  • { skipped: false, droppedPaths: null } — validated, nothing removed.
  • { skipped: false, droppedPaths: [...], instruction } — validated, but those paths were stripped from data: fields the script returned from the API that the outputSchema doesn't declare. If you need them, re-run with output validation skipped.
  • { skipped: true } — validation was bypassed; data is the raw, unchecked script output.

Reading dropped fields / skipOutputDataValidation. To receive the raw, unvalidated result, opt out of output validation (the exact syntax differs by shape — see the reference you loaded above). Input validation is never skipped.

Trimming the result / filterOutputData. To shrink a large result down to the fields you need, pass a jq expression that post-processes data (again, exact syntax per shape). The jq runs against data only, NOT the { data, meta } envelope, so write it rooted at data (run the script's --help — or your shape's equivalent — to see its output schema). The transformed value replaces data, meta is preserved, and the result is NOT re-validated against the output schema.

References

Load the matching reference file before working in that area:

Reference	Covers	Load it when
references/algolia-api-gotchas.md	Auth/ACLs, async indexing (taskID → getTask), objectID-as-string, the record write model + partial-update nested-replace, filtering/faceting DSL, the 1,000-hit pagination cap, rate/size limits, destructive ops, replicas, Recommend, error shape	Before writing records, building filters/facets, configuring an index, or interpreting an error — i.e. almost any non-trivial call