Firestore
Read and write Google Cloud Firestore documents, query collections, and batch documents into a database.
About
Firestore ships in the w6w first-party pack. It declares 12 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.firestore- Version
- 0.1.0
- Author
- w6w
- Licence
- MIT
- Categories
- Databases · Developer Tools
Overview
Firestore gives a workflow a document database: read, create, patch and delete documents by path, list the documents in a collection, and query a collection with equality and range filters, ordering and a limit. Every value is translated to and from Firestore’s typed wire format, so whole numbers stay int64 rather than becoming floats, timestamps and geopoints survive the trip, and a read returns both the raw typed fields and a plain-JS data object.
Beyond single documents, a workflow can fetch many at once and be told which of them do not exist, commit a batch of writes atomically when two documents have to agree, or apply them independently when one failure should not sink the rest. Aggregation queries answer count, sum and average without reading every matching document, and a database’s edition, concurrency mode and location are readable as a connection check.
Collections are addressed the way Firestore addresses them — a path, with the collection id as its own segment — and both a document’s subcollections and the database’s top-level collections can be listed. Health checks track Google Cloud’s own status feed for Cloud Firestore, and declare honestly that Firestore publishes no quota headroom to read.
Build with Firestore
Three routes to the same 12 actions. The Workflow tab is generated from Firestore'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.
List collection ids
collection-list-ids List the collection ids under a document, or the database's top-level collections.
Get the database
database-get Read the Firestore database's metadata: edition, concurrency mode and location.
Create a document
document-create Create a document in a collection, with a chosen or auto-generated id.
Update a document
document-update Patch a document — the fields named by a mask, or the whole document when no mask is given.
Get several documents
documents-batch-get Fetch a list of documents at once, reporting which of them do not exist.
Batch write documents
documents-batch-write Apply many writes independently — each may succeed or fail on its own.
Commit a batch of writes
documents-commit Apply an array of set/update/delete writes atomically, in order.
Run an aggregation query
query-run-aggregation Count, sum or average a collection's fields without reading every document.
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 Firestore's real ids.
{
"manifestVersion": "2",
"name": "firestore-example",
"steps": [
{
"id": "document-create",
"uses": {
"app": "io.w6w.firestore",
"action": "document-create",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"collectionPath": "<collectionPath>",
"data": "<data>"
}
}
]
}Here are some of the things you can do
-
Create a document
performdocument-create -
List collection ids
searchcollection-list-ids -
Get the database
readdatabase-get -
Get a document
readdocument-get -
List documents
searchdocument-list
+7 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 Firestore, 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 Firestore. 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-create",
payload: {
// projectId: "<value>",
// databaseId: "<value>",
collectionPath: "<value>",
// documentId: "<value>",
data: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action document-create --payload '{"collectionPath":"<value>","data":"<value>"}'Give an AI agent Firestore — without giving it Firestore'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.firestore#document-create",
"input": {
"collectionPath": "<collectionPath>",
"data": "<data>"
}
}
}
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 Firestore connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
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.
A tool surface scoped to the caller
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.
A durable workflow in one call
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.
Health-aware discovery
Firestore'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. Firestore itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
Firestore declares its own checks, so its health is a property of the app rather than something the host guesses at.
Firestore platform status
Open incidents for Cloud Firestore on the Google Cloud status dashboard — not the Workspace or Ads one. Unauthenticated and unsigned.