Coda
Read and write Coda docs, tables and rows via the Coda REST API.
About
Coda ships in the w6w first-party pack. It declares 11 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.coda- Version
- 0.1.2
- Author
- w6w
- Licence
- MIT
- Categories
- Productivity · Documents & Files
Overview
Coda puts a doc’s tables and rows into a workflow — list docs, tables and columns, read rows, and write to them with upsert, update or delete. Row writes queue asynchronously and return a request id; a separate action polls that id so a workflow can wait for an edit to actually land before reading the row back, rather than assuming a queued response means the change already happened.
Upsert is the one action worth knowing the shape of: without a set of key columns, every call inserts new rows, so retrying it duplicates them, while providing key columns makes it behave as a true upsert that converges on retry. Update and delete are idempotent either way — reapplying the same cell values, or deleting an already-gone row, lands on the same end state.
Build with Coda
Three routes to the same 11 actions. The Workflow tab is generated from Coda'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.
Get Mutation Status
get-mutation-status Check whether a queued write (from its `requestId`) has been applied.
Get Row
get-row Retrieve a single row. If multiple rows share the identifying column value, an arbitrary one is returned.
List Columns
list-columns List the columns of a table. Returns one page — pass `pageToken` back to walk.
List Docs
list-docs List docs accessible by the connected account. Returns one page — pass `pageToken` back to walk.
List Tables
list-tables List tables (and views) in a doc. Returns one page — pass `pageToken` back to walk.
Update Row
update-row Update cells on an existing row. Queued — returns a `requestId`, not the finished row.
Insert / Upsert Rows
upsert-rows Insert new rows, or upsert (match + update) when `keyColumns` is set. Queued — returns a `requestId`, not the finished rows.
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 Coda's real ids.
{
"manifestVersion": "2",
"name": "coda-example",
"steps": [
{
"id": "get-doc",
"uses": {
"app": "io.w6w.coda",
"action": "get-doc",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"docId": "<docId>"
}
}
]
}Here are some of the things you can do
-
Get Doc
readget-doc -
Get Mutation Status
readget-mutation-status -
Get Row
readget-row -
List Columns
readlist-columns -
List Docs
readlist-docs
+6 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 Coda, 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 Coda. 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: "upsert-rows",
payload: {
docId: "<value>",
tableId: "<value>",
rows: "<value>",
// keyColumns: "<value>",
// disableParsing: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action upsert-rows --payload '{"docId":"<value>","tableId":"<value>","rows":"<value>"}'Give an AI agent Coda — without giving it Coda'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.coda#get-doc",
"input": {
"docId": "<docId>"
}
}
}
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 Coda 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
Coda'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. Coda itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
Coda declares its own checks, so its health is a property of the app rather than something the host guesses at.
Coda platform status
Reads status.coda.io's Atom history feed for open (non-Resolved) incidents.