Search, read and manage Guru cards, collections and folders — the AI-powered knowledge base.
Guru ships in the w6w first-party pack. It declares 16 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.guruGuru is an AI-powered knowledge management platform organized around Cards (individual pieces of verified knowledge), Collections (top-level knowledge bases, usually one per team) and Folders (Guru’s UI calls these “Boards” — the groupings of related Cards within a Collection).
This app covers that core surface headlessly: search Cards by keyword, read a Card’s full content and metadata, create and update Cards (including their verification state), and manage Collections and Folders. It also surfaces team membership, useful for looking up a verifier or owner before assigning one.
Guru’s Trust Score / verification workflow — Cards are marked TRUSTED, STALE or NEEDS_VERIFICATION and carry a verifier who is responsible for keeping them accurate — is a first-class part of the model here: the verify action and the verification fields on every Card action are not an afterthought bolted on top of a generic CMS shape.
Guru’s newer AI Agent / Knowledge Agent / Chat surface (agents, skills, quality runs, MCP configuration) is a separate and much larger subsystem and is deliberately out of scope for this app; see the README.
Three routes to the same 16 actions. The Workflow tab is generated from Guru'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.
card-search Full-text search over Cards. Returns at most 50 per call — page with nextToken.
card-update Replace a Card's title, content, share status and Folder placement. Does not touch tags.
card-update-content Update a Card's body. Guru's own schema requires the title alongside the content on this call — see the action's own notes.
member-list List the team's members — useful for looking up a verifier or owner by name.
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 Guru's real ids.
{
"manifestVersion": "2",
"name": "guru-example",
"steps": [
{
"id": "card-create",
"uses": {
"app": "io.w6w.guru",
"action": "card-create",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"title": "<title>",
"content": "<content>"
}
}
]
}card-create card-get card-search card-update card-update-content +11 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 Guru, 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 Guru. 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: "card-create",
payload: {
title: "<value>",
content: "<value>",
// collectionId: "<value>",
// folderIds: "<value>",
// shareStatus: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action card-create --payload '{"title":"<value>","content":"<value>"}' Give an AI agent Guru — without giving it Guru'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.guru#card-create",
"input": {
"title": "<title>",
"content": "<content>"
}
}
}
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 Guru 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.
Guru'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. Guru itself is MIT, and the runtime that executes it is source-available (FSL).
Guru declares its own checks, so its health is a property of the app rather than something the host guesses at.
Component status from status.getguru.com. Covers the API, web app, servers, extension, Slack bot, file service and analytics, plus the Stripe/Zuora billing infrastructure Guru itself depends on.
No machine-readable surface: Guru's OpenAPI document declares no response headers, and none were observed live on any endpoint.