Manage Lokalise projects, keys, translations, languages, files, contributors, tasks and webhooks.
Lokalise ships in the w6w first-party pack. It declares 31 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.lokaliseLokalise is a translation and localization management platform: strings live as keys with per-language translations, organized into projects, exported as files in the format each platform (iOS, Android, web, …) expects.
This app covers the path a workflow actually needs around that: manage projects and their languages, create and update keys and translations, upload a source file for Lokalise to parse and queue for translation, and export a project’s current translations as a downloadable bundle. Beyond the core content, it covers contributors (who can translate what), comments (in-context notes on a key), tasks (assigning translation work with a due date) and webhooks (reacting to a translation event instead of polling for it).
A structural fact shapes every ‘create’ action here: Lokalise’s create endpoints are bulk-only — even creating one key, language, contributor or comment means POSTing an array of one, and a 200 response can still carry a per-item ‘errors’ array for partial failure. Every bulk action in this app surfaces that array rather than assuming HTTP success means every item succeeded.
Three routes to the same 31 actions. The Workflow tab is generated from Lokalise'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.
contributor-list List the project's contributors and their per-language permissions.
file-delete Delete a file and every key attributed to it. Documents and Marketing projects only — not supported on Software (localization_files) projects.
file-download Export the project's current translations as a downloadable .zip bundle.
language-list-system List every language Lokalise supports, independent of any project.
translation-list List all translation items in a project, one row per key/language pair.
translation-update Set a translation's text and/or its unverified/reviewed flags.
webhook-create Subscribe a URL to project events. Not deduplicated — a retry creates a second webhook.
webhook-list List the project's webhooks. The result includes each webhook's signing secret — treat it as sensitive.
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 Lokalise's real ids.
{
"manifestVersion": "2",
"name": "lokalise-example",
"steps": [
{
"id": "comment-create",
"uses": {
"app": "io.w6w.lokalise",
"action": "comment-create",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"projectId": "<projectId>",
"keyId": "<keyId>",
"comments": "<comments>"
}
}
]
}comment-create comment-list contributor-create contributor-list file-download +26 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 Lokalise, 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 Lokalise. 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: "comment-create",
payload: {
projectId: "<value>",
keyId: "<value>",
comments: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action comment-create --payload '{"projectId":"<value>","keyId":"<value>","comments":"<value>"}' Give an AI agent Lokalise — without giving it Lokalise'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.lokalise#comment-create",
"input": {
"projectId": "<projectId>",
"keyId": "<keyId>",
"comments": "<comments>"
}
}
}
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 Lokalise 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.
Lokalise'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. Lokalise itself is MIT, and the runtime that executes it is source-available (FSL).
Lokalise declares its own checks, so its health is a property of the app rather than something the host guesses at.
Component status from status.lokalise.com: Lokalise.com, Lokalise API, Lokalise App, Lokalise OTA and Lokalise Messages.
Users, keys, projects, OTA traffic and AI words consumed vs. allocated, per team, read from GET /teams.
Remaining requests in Lokalise's per-token rate-limit window, read from the x-ratelimit-remaining response header on a List Projects call.