Set up WorkerKit in Claude.

Before you start

What you need.

A WorkerKit account, and admin access on it: choosing the key a connection will hold is an admin-only decision, on the consent page and at Fleet access alike. The Free plan is enough, and no card is needed.

Any Claude surface that adds a custom connector or an MCP server: claude.ai, Claude Desktop, Cowork or Claude Code. The terminal is the one place a key can go on a header instead of through the sign-in flow.

About a minute of the human's time. Claude does everything except approving the connection.

Why this route: Claude reaches remote MCP servers natively on every surface, so it takes the hosted WorkerKit MCP mounts directly, with nothing to install and no code to write. The fleet mount publishes OAuth discovery and accepts dynamic client registration, so on the hosted surfaces the connection is a URL and a sign-in rather than a credential anyone handles.

One address is worth reading twice: workerkit.ai/mcp is the documentation page, and the live mounts are on mcp.workerkit.ai. The table above has both in full.

The case for connecting at all is on the Claude page.

Step 0. Human and agent, under a minute

Add the public catalog.

The catalog mount takes no credential at all, so it connects on any Claude surface and proves the wiring before any key is involved.

• Claude Code claude mcp add --transport http workerkit-directory https://mcp.workerkit.ai/directory
• claude.ai, Claude Desktop and Cowork Customize > Connectors > + > Add custom connector URL: https://mcp.workerkit.ai/directory Authentication: No sign-in

The catalog mount takes no credential at all, on either path. It is the fastest proof that the client can reach WorkerKit, and it is worth adding first for exactly that reason: if it connects and the fleet mount does not, the problem is the credential rather than the wiring.

On Team and Enterprise only an Owner can add a custom connector, from Organization settings, then Connectors, then Add, then Custom; members then Connect it individually afterwards. A Free plan holds one custom connector, so on Free this is the one to spend it on only if the fleet mount goes into Claude Code instead.

Step 1. Human, about 60 seconds

Connect your fleet.

Add the fleet mount by URL and let Claude run the OAuth flow, or pass a manager key on one header in Claude Code. Either way the reach is exactly the scopes on the key.

The fleet mount is an OAuth 2.1 authorization server as well as an MCP server, so on claude.ai, Claude Desktop and Cowork it is added exactly like the catalog: paste the URL and approve. Nothing is typed into the chat, and there is no beta to wait for.

1. 01

   Give the client the URL and nothing else

   The fleet mount publishes OAuth discovery, so a client that supports it finds the authorization server on its own and registers itself. There is no client ID to create, no secret to store and no redirect address to register in advance.

2. 02

   Approve on the WorkerKit consent page

   The client sends the person to https://workerkit.ai/mcpauth. Signing in there is what proves who is connecting; nothing is typed into the assistant.

3. 03

   An account admin picks the key the client will hold

   The consent page lists the account's manager keys and can create one on the spot, so a first connect needs no visit to Fleet access beforehand. It is admin-only on the server, not just in the page: a member sees an honest notice rather than a broken screen.

4. 04

   Scopes are chosen with the key, and are the whole of the reach

   Whatever that key carries is exactly what the assistant can do, and nothing on it reaches a connected app. Narrow it now or widen it later; a new key drops into the same connection.

• claude.ai, Claude Desktop and Cowork: add it and approve Customize > Connectors > + > Add custom connector URL: https://mcp.workerkit.ai/workers Then approve at https://workerkit.ai/mcpauth
• Claude Code, with a key created at Fleet access claude mcp add --transport http workerkit-workers https://mcp.workerkit.ai/workers --header "Authorization: Bearer pe_mgr_..."

Nothing is pasted into the chat on this path, which is its real advantage: the credential never passes through a message, a log or a screenshot, and revoking it later is one action in the dashboard rather than an edit inside the assistant.

A client that cannot run an OAuth flow takes the same fleet on one header instead. An account admin creates a key at https://workerkit.ai/fleet-access, copies it once, and it becomes the value of an Authorization header on the fleet mount. That header is the whole of the authentication: there is nothing to register and nothing to pre-approve. In Claude Code that is the --header flag above, which is generally available and takes the key as it is. Scope flags come before the server name: --scope local is the default, --scope user puts it in every project of yours, and --scope project writes a .mcp.json the repository can carry, whose headers value should be an environment reference rather than the key itself.

There is a third path on claude.ai, and it is the one to reach for last: a Request header entered in the Add custom connector dialog. Anthropic documents that field as beta and open to a limited set of organizations, its value is sent exactly as typed (so it reads Bearer pe_mgr_..., scheme and space included), and authentication settings cannot be edited after a connector is added, which turns a key rotation into a remove and re-add. Use it only where OAuth is not wanted.

A key carries the scopes it was granted when it was created. Granting a scope later never reaches an existing key: if a call answers 403 OPERATION_NOT_ALLOWED, an admin re-scopes the key at Fleet access or creates a new one, and the new key drops into the same place. Read key_info rather than assuming.

Step 2. Agent

Verify the scopes.

Call key_info on the fleet mount. The scope list it returns is the truth, whatever was intended.

• Ask the key what it can do (no arguments) key_info

It returns the account title, the key name, the key's scopes and expiry, and serverTimeUtc; any valid key may ask. Confirm the scope list matches what was granted. If anything is missing it was either not granted, or granted after the key was created, and an admin goes back to Fleet access to re-scope or create a new key. Plan against what this call returns rather than what was intended.

Step 3. Agent

Keep the WorkerKit skill.

Save the skill file where this Claude surface keeps skills, or re-read the URL at the start of any session that touches the fleet.

• Claude Code: write it as a skill folder, personal or project ~/.claude/skills/workerkit-fleet/SKILL.md .claude/skills/workerkit-fleet/SKILL.md Source: https://workerkit.ai/claude/SKILL.md

The file is a standard skill: a name and a description in its front matter, then the tool reference, the cost-field semantics and the safety rules the agent needs. In Claude Code a skill is a folder on disk and is discovered at startup, so writing the file into one of the paths above is the whole install. On claude.ai the same content is uploaded as a ZIP containing workerkit-fleet/SKILL.md under Customize, then Skills, which first needs code execution and file creation enabled, and on Team and Enterprise skills enabled for the organization too.

A surface that keeps no skills loses nothing: fetch the same URL at the start of any session that touches the fleet and read it before the first call. It is one request, the content is identical, and it is always the current version.

Step 4. Agent

Run the smoke test.

List the fleet and read recent runs, which alone prove the setup. If the account has workers, read one before running it, run it with approval, and report the receipt.

If the fleet is empty, stop after step 3 and say so: a new account has no workers yet, and the steps above have already proved the setup, because the key authenticated and both read scopes answered. Do not read it as a fault, and do not invent a worker to run. Go to the first-worker step and give the fleet something to do.

If the fleet has workers, pick one whose effects are reads. Judge it from the instruction and the permissions, never from the name: a worker whose apps are all read-level and whose instruction only gathers, summarises or reports is safe to run; one that sends, posts, writes, orders or pays is not, whatever it is called. If none is clearly read-only, run one with an explicit prompt for that run bounding it to a read.

• 1. The directory is alive: confirm a kit count comes back directory_overview
• 2. The fleet list: report how many workers the assistant can see workers_list
• 3. Recent runs across the fleet: confirms readRuns runs_feed { "limit": 5 }
• 4. Before running anything, read what that worker does and what it may touch. Both ride on readWorkers, and this is the step that decides whether the run is safe instruction_get worker_permissions_get
• 5. Run it with the person's approval, then poll every 3 to 5 seconds until a terminal event arrives: run.succeeded, run.failed, run.blocked or run.awaiting_input worker_run run_events { "afterSeq": <lastSeq> }
• 6. Read the receipt: costAuthority first, then modelCostUsd run_get

Read costAuthority first, then modelCostUsd beside it, and report both. A null finalDigest is not a failure: a run produces one only when its harness wrote a report. A costAuthority of unbilled_estimate, none or in_flight is not a failure either; it says how far to trust the number, and the operating rules below say what each value means.

A Skipped run is also a pass, and it is the cost brake working: the gate refused the run before any money moved. Report the skipReason rather than retrying. Then report the fleet size, the active scopes, and what the test cost, and the fleet is live.

Step 5. Human and agent

Give the fleet its first worker.

A new account starts empty. Install a kit from its page in one click, or let Claude preview and install one with kit_install_preview and kit_install once the person approves.

A new WorkerKit account starts with no workers, so the last step of setup is getting one. A kit is a ready-made worker: its instruction, its exact app permissions and its schedule, written and reviewed in advance, so installing one is how a fleet usually starts rather than a shortcut. There are two paths, and either ends with something to operate.

1. 01

   The one-click path: install a kit from its page

   Browse the catalog at workerkit.ai/kits, open a kit whose job you recognise, and install it. It arrives pre-instructed, pre-permissioned and pre-scheduled, and asks for whatever it needs to connect. This path needs no scope on the manager key at all, so it works on the first session whatever the key can do.

2. 02

   The agent path: shortlist, preview, approve, install

   With installKits on the key, the assistant can do the whole thing in chat: shortlist with kits_search, read the candidate with kit_get, then call kit_install_preview to surface its setup questions and exactly what the new worker would be allowed to touch. Put that in front of the person in plain words, and call kit_install once they approve, with deploy set to true so the new worker is actually able to run (that half needs manageDeployments on the key; without it the worker is installed but inert until worker_deploy). The new worker's key is shown once in the response.

3. 03

   Then run it once and read the receipt

   A worker earns its place on the first run. Trigger it with worker_run, follow run_events until a terminal event arrives, and report the digest and the settled cost from run_get. Set a schedule with schedule_create when the person is happy with what it did, and the fleet is not just connected, it is working.

The catalog is public, so Claude can read the whole of it through the directory mount before anyone signs in: it can shortlist candidates and explain what each one would do while the person decides.

Classification on demand

Create a decision worker from your questions.

Categorize, score or triage connected data, then reuse and adjust the worker.

Reuse an existing worker or decision kit when its questions fit. For a custom classification task, creation compiles a supported source recipe and typed questions into a private kit and an installed decision worker. Optional deploy:true adds deployment; creation never starts a run or adds a schedule. The kit can be edited and published later through the normal kit lifecycle.

On the workers connection, call kit_app_tools(purpose:"decision") for recipes, argument schemas, permissions and examples, then kit_authoring_guide(section:"decision"). Check account connections with apps_list; public discovery does not check them. Send the request to decision_worker_create, follow nextCall, then use worker_run and run_get. Running needs runWorkers; results need readRuns. Use instruction_get / instruction_set for saved category or level answers, or answers on one run. Structural question/source changes need a revised kit and a replacement install. Check receipt status, coverage, omissions, warnings and cost before claiming completion.

Start with email-previews, calendar-events or sheets-rows. These recipes return judgments without app writes. Email previews do not include full threads; calendar events are invitation data, not transcripts; Sheets needs a Google file ID, a finite tab-qualified range and a columns map. A connected app alone does not make every tool a supported source. Source filters select evidence, not permissions: the Sheets recipe grants spreadsheet reads across the linked Drive account.

Supply 1–8 questions: choice with named options, score with ordered levels, or noul for the probability of a statement. Choice adds unclear automatically. Set an explicit confidenceFloor between 0 and 1; it routes uncertainty and does not promise accuracy. maxItems is 1–50, default 20. Creation requires publishKits and installKits; optional deployment also needs manageDeployments.

Use a fresh requestId for each new worker. Retry with the same ID and identical body to recover the original receipt; a changed body returns 409. The response includes tokenId for MCP/CLI worker commands, stable workerId, kitSlug, readiness and nextCall. A deploymentError means the worker already exists: fix deployment on that worker. The receipt is a snapshot; check current worker readiness before running. No worker API secret is returned.

The workers connection includes discovery and authoring, so classification needs one MCP connection. Its full profile has 91 tools (84 manager tools and 7 public discovery reads); only decision_worker_create is new. For a smaller list, a client that supports custom headers can send X-WorkerKit-Profile: decision on every request, including initialization, listing and calls, to select 20 workflow tools. A server operator can instead set MCP_WORKERS_PROFILE=decision. The endpoint and existing sign-in stay the same; a profile grants no additional permissions. Reconnect and relist after changing profiles. Use the advertised tool list: fleet administration, publishing and event-stream tools require the full profile. Successful responses include structured content and a text fallback; creation declares an output schema.

Read the request example and full contract.

Least privilege

What each scope grants.

The key's scopes are exactly its reach, and a refusal names the scope it wanted. Start with the reads and add scopes as the work asks for them.

Operating rules

What the agent follows on every call.

The same rules the skill file carries, so a session that read neither still has them. Every row is a fact of the platform, not advice.

Quick answers

Symptoms, causes, fixes.

Each WorkerKit failure names itself in the response, so an agent never has to guess which one it met: read the code, apply the row, and carry on.