Search, index documents, and manage Meilisearch indexes and their settings.
Meilisearch ships in the w6w first-party pack. It declares 24 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.meilisearchMeilisearch is a fast, self-hostable search engine, and this app runs both the search and the indexing side of it: search an index with filters, sorting and facets, run several searches at once, add or delete documents, and manage indexes and their settings.
Because a Meilisearch instance can be a customer’s own server as easily as a cloud project, this app connects to whichever address a workflow’s credentials name rather than a fixed vendor host. Writes are asynchronous — adding a document or changing settings returns a task immediately, before the work has actually happened — so this app returns that task and gives a workflow a way to check whether it later succeeded, rather than treating an immediate 200 response as confirmation.
Beyond search and documents, this app manages tasks (checking status, cancelling pending ones), reads API keys and their scopes, and reports index and database statistics. Backup, snapshotting, and minting new API keys are left out, since those carry consequences a workflow step shouldn’t trigger casually.
Three routes to the same 24 actions. The Workflow tab is generated from Meilisearch's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
document-add Enqueue documents into an index, replacing or merging by primary key.
documents-clear Enqueue deletion of every document in an index. The index and settings remain.
index-stats Document count, whether the index is still indexing, and field distribution.
index-update Enqueue a primary key change. Only succeeds while the index is empty.
settings-reset Enqueue a reset of EVERY setting on an index back to its default.
settings-update Enqueue a settings change. Unset fields are left alone; changes re-index.
similar-documents Recommend documents similar to one you name. Requires a configured embedder.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Meilisearch's real ids.
{
"manifestVersion": "2",
"name": "meilisearch-example",
"steps": [
{
"id": "document-add",
"uses": {
"app": "io.w6w.meilisearch",
"action": "document-add",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"documents": "<documents>",
"mode": "<mode>"
}
}
]
}document-add document-get document-list documents-clear facet-search +19 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Meilisearch, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Meilisearch. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "document-add",
payload: {
// indexUid: "<value>",
documents: "<value>",
mode: "<value>",
// primaryKey: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action document-add --payload '{"documents":"<value>","mode":"<value>"}' Give an AI agent Meilisearch — without giving it Meilisearch's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.meilisearch#document-add",
"input": {
"documents": "<documents>",
"mode": "<mode>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Meilisearch connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Meilisearch's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Meilisearch itself is MIT, and the runtime that executes it is source-available (FSL).
Meilisearch declares its own checks, so its health is a property of the app rather than something the host guesses at.
This connection's own server, via its unauthenticated /health endpoint. Sends no credential — an expired key must not make the server look down.
Open incidents on Meilisearch Cloud's status feed. Speaks only for Cloud — a self-hosted instance is covered by the `instance` check instead.