Algolia

Algolia Agent Studio API

Algolia's agent-building runtime: agents, conversations, tools, memory, guardrails and per-turn context, exposed as 42 REST operations. Notably, Agent Studio can itself consume third-party MCP tools, so Algolia is both an MCP server and an MCP client.

GitHub OpenAPI

Operations 42

Each operation below carries the questions people ask an LLM about it and the instructions they give an agent to run it. Generated by API Evangelist overlay

GET /1/agents List Agent Studio agents · List Agents #
Ask an LLM
“Which AI agents have I built in Algolia Agent Studio?”
“Can I list only the agents that use a particular LLM provider?”
Tell an agent
List all my Agent Studio agents.
Show agents that use LLM provider {providerId}, page {page}.
POST /1/agents Create a new AI agent · Create Agent #
Ask an LLM
“How do I create a new conversational agent with its own instructions?”
“Can I pick the model and attach tools when I first create an agent?”
Tell an agent
Create an agent named {name} with instructions {instructions}.
Create agent {name} on provider {providerId} using model {model}, with instructions {instructions} and tools {tools}.
GET /1/agents/{agentId} Get an agent's details · Get Agent #
Ask an LLM
“How do I see the instructions, model and tools one agent is configured with?”
“What settings does a specific agent have right now?”
Tell an agent
Show agent {agentId}.
Retrieve the full configuration of agent {agentId}.
PATCH /1/agents/{agentId} Update an existing agent · Update Agent #
Ask an LLM
“How do I change the system prompt of an agent I already built?”
“Can I switch an existing agent to a different model without recreating it?”
Tell an agent
Update agent {agentId} to use model {model}.
Change the instructions of existing agent {agentId} to {instructions}.
DELETE /1/agents/{agentId} Delete an agent · Delete Agent #
Ask an LLM
“How do I permanently remove an agent I no longer need?”
“Can I delete an agent by its ID?”
Tell an agent destructive · confirm first
Delete agent {agentId}.
Remove agent {agentId} from Agent Studio.
GET /1/agents/{agentId}/allowed-domains List an agent's allowed domains · List Allowed Domains #
Ask an LLM
“Which websites are allowed to call my agent?”
“How do I see every domain pattern whitelisted for one agent?”
Tell an agent
List the allowed domains for agent {agentId}.
Show every domain pattern agent {agentId} accepts requests from.
POST /1/agents/{agentId}/allowed-domains Allow one domain to use an agent · Create Allowed Domain #
Ask an LLM
“How do I let my app's domain call my agent?”
“Can I allow a wildcard subdomain pattern like *.example.com for an agent?”
Tell an agent
Allow domain {domain} for agent {agentId}.
Add the single domain pattern {domain} to agent {agentId}'s allowlist.
POST /1/agents/{agentId}/allowed-domains/bulk Allow many domains for an agent at once · Bulk Insert Allowed Domains #
Ask an LLM
“How do I add a whole list of domains to an agent's allowlist in one go?”
“What happens to duplicates when I bulk-add allowed domains?”
Tell an agent
Bulk add domains {domains} to agent {agentId}'s allowlist.
Allow all of these domains for agent {agentId}, skipping duplicates: {domains}.
DELETE /1/agents/{agentId}/allowed-domains/bulk Remove many allowed domains from an agent · Bulk Delete Allowed Domains #
Ask an LLM
“How do I remove several domains from an agent's allowlist at once?”
“Can I bulk-revoke allowed domains by their IDs?”
Tell an agent destructive · confirm first
Remove allowed domains {domainIds} from agent {agentId}.
Bulk delete these domain IDs from agent {agentId}'s allowlist: {domainIds}.
GET /1/agents/{agentId}/allowed-domains/{domainId} Get one allowed domain for an agent · Get Allowed Domain #
Ask an LLM
“How do I look up a single allowed-domain entry by its ID?”
“What pattern is stored for one specific allowed domain on my agent?”
Tell an agent
Show allowed domain {domainId} on agent {agentId}.
Retrieve the domain pattern for entry {domainId} of agent {agentId}.
DELETE /1/agents/{agentId}/allowed-domains/{domainId} Remove one allowed domain from an agent · Delete Allowed Domain #
Ask an LLM
“How do I stop one domain from calling my agent?”
“Can I remove a single allowed domain by its ID?”
Tell an agent destructive · confirm first
Remove allowed domain {domainId} from agent {agentId}.
Revoke the single domain entry {domainId} on agent {agentId}.
DELETE /1/agents/{agentId}/cache Clear an agent's cached completions · Invalidate Agent Cache #
Ask an LLM
“How do I clear cached answers after I change my agent's data?”
“Can I invalidate only the completions cached before a certain date?”
Tell an agent destructive · confirm first
Invalidate the completion cache for agent {agentId}.
Clear agent {agentId}'s cached completions created before {before}.
POST /1/agents/{agentId}/completions Send a message to an agent and get a reply · Create Completion #
Ask an LLM
“How do I chat with my agent and get its response?”
“Can I stream an agent's answer back token by token?”
Tell an agent
Send messages {messages} to agent {agentId} in compatibility mode {compatibilityMode}.
Stream a reply from agent {agentId} to {messages} with compatibility mode {compatibilityMode} and stream={stream}.
GET /1/agents/{agentId}/conversations List an agent's conversations · List Conversations #
Ask an LLM
“What conversations have users had with my agent this week?”
“Can I see only the conversations that got a thumbs-down vote?”
Tell an agent
List conversations for agent {agentId} between {startDate} and {endDate}.
Show agent {agentId}'s conversations with feedback vote {feedbackVote}.
DELETE /1/agents/{agentId}/conversations Delete an agent's conversations in a date range · Delete Conversations #
Ask an LLM
“How do I purge all of an agent's conversation history for a period?”
“Can I bulk-delete an agent's conversations older than a certain date?”
Tell an agent destructive · confirm first
Delete all conversations for agent {agentId}.
Delete agent {agentId}'s conversations from {startDate} to {endDate}.
GET /1/agents/{agentId}/conversations/export Export an agent's conversations · Export Conversations #
Ask an LLM
“How do I download every conversation my agent has had for offline analysis?”
“Can I export an agent's conversations for just last month?”
Tell an agent
Export all conversations for agent {agentId}.
Export agent {agentId}'s conversations between {startDate} and {endDate}.
GET /1/agents/{agentId}/conversations/{conversationId} Get one conversation and its messages · Get Conversation #
Ask an LLM
“How do I read the full transcript of one agent conversation?”
“Can I include user feedback when I pull up a single conversation?”
Tell an agent
Show conversation {conversationId} of agent {agentId}.
Retrieve the messages and feedback of conversation {conversationId} for agent {agentId}.
DELETE /1/agents/{agentId}/conversations/{conversationId} Delete one conversation · Delete Conversation #
Ask an LLM
“How do I delete a single conversation from an agent's history?”
“Can I remove one specific chat by its conversation ID?”
Tell an agent destructive · confirm first
Delete conversation {conversationId} of agent {agentId}.
Remove the single chat {conversationId} from agent {agentId}.
POST /1/agents/{agentId}/publish Publish an agent · Publish Agent #
Ask an LLM
“How do I make my agent live after I finish configuring it?”
“What's the step to take a draft agent into production?”
Tell an agent
Publish agent {agentId}.
Take agent {agentId} live.
POST /1/agents/{agentId}/unpublish Unpublish an agent · Unpublish Agent #
Ask an LLM
“How do I take a published agent offline without deleting it?”
“Can I pull an agent out of production temporarily?”
Tell an agent
Unpublish agent {agentId}.
Take agent {agentId} offline but keep its configuration.
GET /1/configuration Get the Agent Studio configuration · Get Configuration #
Ask an LLM
“How long does Agent Studio keep my conversation data?”
“What application-level Agent Studio settings are in effect?”
Tell an agent
Show my Agent Studio configuration.
Tell me the current data retention setting for Agent Studio.
PATCH /1/configuration Change Agent Studio data retention · Patch Configuration #
Ask an LLM
“How do I shorten how many days Agent Studio retains data?”
“Can I change the maximum retention period for conversations?”
Tell an agent
Set Agent Studio's maximum retention to {maxRetentionDays} days.
Keep Agent Studio data for no more than {maxRetentionDays} days.
POST /1/feedback Record feedback on an agent message · Create Feedback #
Ask an LLM
“How do I capture a thumbs-up or thumbs-down on an agent's answer?”
“Can I attach tags and notes to feedback on a specific message?”
Tell an agent
Record a {vote} vote on message {messageId} from agent {agentId}.
Leave feedback {vote} with notes {notes} on agent {agentId}'s message {messageId}.
GET /1/providers List configured LLM providers · List Providers #
Ask an LLM
“Which LLM providers have I connected to Agent Studio?”
“Can I page through my configured model providers?”
Tell an agent
List my configured LLM providers.
Show page {page} of my LLM providers, {limit} per page.
POST /1/providers Connect a new LLM provider · Create Provider #
Ask an LLM
“How do I connect my own LLM provider account so agents can use it?”
“What do I need to supply to register a model provider in Agent Studio?”
Tell an agent
Add LLM provider {providerName} as {name} with settings {input}.
Register a new {providerName} connection named {name} using credentials {input}.
GET /1/providers/models List models available across providers · Get Provider Models #
Ask an LLM
“Which LLM models can my agents choose from?”
“What models does Agent Studio support overall?”
Tell an agent
List all models available to Agent Studio.
Show every model my agents could be configured with.
GET /1/providers/{providerId} Get one LLM provider's details · Get Provider #
Ask an LLM
“How do I view the settings of one connected LLM provider?”
“What's stored for a specific provider connection?”
Tell an agent
Show LLM provider {providerId}.
Retrieve the configuration of provider connection {providerId}.
PATCH /1/providers/{providerId} Update an LLM provider connection · Update Provider #
Ask an LLM
“How do I rotate the credentials on an existing LLM provider connection?”
“Can I rename a provider connection?”
Tell an agent
Update provider {providerId} with new settings {input}.
Rename provider connection {providerId} to {name}.
DELETE /1/providers/{providerId} Remove an LLM provider connection · Delete Provider #
Ask an LLM
“How do I disconnect an LLM provider from Agent Studio?”
“Can I delete a provider connection I no longer pay for?”
Tell an agent destructive · confirm first
Delete LLM provider {providerId}.
Disconnect provider connection {providerId} from Agent Studio.
GET /1/providers/{providerId}/models List models offered by one provider · Get Provider Models By Id #
Ask an LLM
“Which models can I use through a specific provider connection?”
“What models does one of my connected LLM providers expose?”
Tell an agent
List the models available from provider {providerId}.
Show which models provider connection {providerId} offers.
GET /1/secret-keys List Agent Studio secret keys · List Secret Keys #
Ask an LLM
“Which secret keys have been created for my agents?”
“Can I page through my Agent Studio secret keys?”
Tell an agent
List my Agent Studio secret keys.
Show page {page} of secret keys, {limit} at a time.
POST /1/secret-keys Create a secret key for agents · Create Secret Key #
Ask an LLM
“How do I generate a secret key to call my agents from a backend?”
“Can I scope a new secret key to specific agents only?”
Tell an agent
Create a secret key named {name}.
Create secret key {name} limited to agents {agentIds}.
GET /1/secret-keys/{secretKeyId} Get one secret key's details · Get Secret Key #
Ask an LLM
“How do I see which agents one secret key is scoped to?”
“What details are stored for a specific Agent Studio secret key?”
Tell an agent
Show secret key {secretKeyId}.
Retrieve the name and agent scope of secret key {secretKeyId}.
PATCH /1/secret-keys/{secretKeyId} Update a secret key's name or agent scope · Patch Secret Key #
Ask an LLM
“How do I change which agents an existing secret key can access?”
“Can I rename a secret key after creating it?”
Tell an agent
Scope secret key {secretKeyId} to agents {agentIds}.
Rename secret key {secretKeyId} to {name}.
DELETE /1/secret-keys/{secretKeyId} Delete a secret key · Delete Secret Key #
Ask an LLM
“How do I revoke a leaked Agent Studio secret key?”
“Can I delete a secret key by its ID?”
Tell an agent destructive · confirm first
Delete secret key {secretKeyId}.
Revoke Agent Studio secret key {secretKeyId} permanently.
GET /1/user-data/{userToken} Get all stored data for a user token · Get Data By User Token #
Ask an LLM
“How do I retrieve every memory and conversation Agent Studio holds for one end user?”
“Can I pull a user's full data for a data access request?”
Tell an agent
Get all memories, conversations and messages for user token {userToken}.
Export everything Agent Studio stores about user {userToken}.
DELETE /1/user-data/{userToken} Delete all messages for a user token · Delete Data By User Token #
Ask an LLM
“How do I permanently erase an end user's agent messages?”
“Does deleting a user's data also remove their conversations?”
Tell an agent destructive · confirm first
Permanently delete all messages for user token {userToken}.
Erase user {userToken}'s agent messages, leaving the conversations in place.
GET /{path} Send a raw GET request to the Agent Studio API · Send requests to the Algolia REST API #
Ask an LLM
“Can I read an Agent Studio endpoint that the client library doesn't wrap?”
“Is there a generic way to GET any Agent Studio path with custom query parameters?”
Tell an agent
Send a raw GET request to Agent Studio path {path}.
Fetch Agent Studio path {path} with query parameters {parameters}.
POST /{path} Send a raw POST request to the Agent Studio API · Send requests to the Algolia REST API #
Ask an LLM
“Can I POST to an Agent Studio endpoint the SDK doesn't expose as a method?”
“What's the generic escape hatch for posting to any Agent Studio path?”
Tell an agent
Send a raw POST request to Agent Studio path {path}.
POST to Agent Studio path {path} with query parameters {parameters}.
PUT /{path} Send a raw PUT request to the Agent Studio API · Send requests to the Algolia REST API #
Ask an LLM
“Can I send a custom PUT to an Agent Studio path that has no dedicated method?”
“Is there a passthrough for PUT calls on an arbitrary Agent Studio endpoint?”
Tell an agent
Send a raw PUT request to Agent Studio path {path}.
PUT to Agent Studio path {path} with query parameters {parameters}.
DELETE /{path} Send a raw DELETE request to the Agent Studio API · Send requests to the Algolia REST API #
Ask an LLM
“Can I issue a DELETE against an arbitrary Agent Studio path?”
“Is there a generic delete call for Agent Studio endpoints without a dedicated method?”
Tell an agent destructive · confirm first
Send a raw DELETE request to Agent Studio path {path}.
DELETE Agent Studio path {path} with query parameters {parameters}.
GET /setClientApiKey Switch the API key the Agent Studio client uses · Switch the API key used to authenticate requests #
Ask an LLM
“Can I change which API key my Agent Studio client authenticates with?”
“How do I rotate the Algolia API key an existing Agent Studio client sends?”
Tell an agent
Switch the Agent Studio client to authenticate with API key {apiKey}.
Use {apiKey} for all further Agent Studio API requests.

Documentation

📖
Documentation
https://www.algolia.com/doc/guides/algolia-ai/agent-studio

Specifications

⚙
OpenAPI
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/openapi/algolia-agent-studio-api-openapi.yml

Schemas & Data

📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/json-schema/algolia-agent-config-update-schema.json
📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/json-schema/algolia-agent-config-create-schema.json
📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/json-schema/algolia-conversation-full-response-schema.json
📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/json-schema/algolia-agent-with-version-response-schema.json
📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/json-schema/algolia-feedback-response-schema.json

Other Resources

🔗
Overlay
https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/overlays/algolia-agent-studio-api-overlay.yaml

Work with this as data

Every API here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for apis

7 MCP tools reach this
  • find_apisBrowse and filter every API in the catalog.
  • get_api_artifactsOne API's artifacts, grouped by type.
  • get_openapiThe primary OpenAPI for this API.
  • find_similar_apisAPIs that look like this one.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This API 
curl "https://apis.io/api/v1/apis/algolia:algolia-agent-studio-api"
All apis 
curl "https://apis.io/api/v1/apis?limit=25"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.

GitHub Google LinkedIn

A second provider on the same verified email joins the account you already have.

OpenAPI Specification

algolia-agent-studio-api-openapi.yml Raw ↑ 
openapi: 3.2.0
info:
  title: Agent Studio API
  description: The Agent Studio API lets you build and operate generative AI agents that use Algolia data, tools, and your chosen LLM provider.
  version: 0.1.0
servers:
- url: https://{APPLICATION_ID}.algolia.net/agent-studio
  description: Agent Studio API.
  variables:
    APPLICATION_ID:
      default: EXAMPLE
      description: Your Algolia application ID.
security:
- appId: []
  apiKey: []
tags:


# --- truncated at 32 KB (116 KB total) ---
# Full source: https://raw.githubusercontent.com/api-evangelist/algolia/refs/heads/main/openapi/algolia-agent-studio-api-openapi.yml