PocketDocs

Creating an agent

Give an agent a wallet, then connect the agent you already run.

An agent is whatever you already use to do the work. Pocket gives it a wallet and a set of limits.

Create it

Go to Agents and choose New agent. Give it a name, and optionally a line about what it does.

A wallet is created at the same time, and the address appears on the agent's page.

Your new agent cannot spend anything yet. It stays that way until you give it a budget, which is the next page.

Fund the wallet

Copy the address from the agent's page and send it some stablecoin. The Connect agent screen links to a faucet if you are on testnet.

The wallet is already set up to hold the token, so there is nothing else to do. The balance appears once the transfer confirms.

Connect the agent you run

Open Connect agent in the dashboard and copy the line for whichever agent you use.

hermes mcp add pocket --transport http --url https://your-mcp-url/mcp

There is no key in any of these, on purpose.

Your agent never holds the key

The key lives in your own server's environment. It is never passed to your agent, never returned in a result, and never written into the model's context. An agent that cannot read it cannot leak it.

What it can do once connected

Eight tools reach your agent. One can spend, and the other seven exist so it does not have to guess.

ToolWhat it doesSpends
pocket_pay_for_resourceBuys something and returns the contentYes
pocket_preview_paymentAsks whether a purchase would be allowedNo
pocket_create_task_budgetOpens a ring-fenced amount for one jobNo
pocket_list_agentsLists agents with budgets and spendNo
pocket_get_agentOne agent, with its limits and balanceNo
pocket_spend_summarySpend by category and sellerNo
pocket_list_paymentsRecent purchases, refusals includedNo
pocket_audit_trailThe full recordNo

Pausing and revoking

On the agent's page you can pause it, which refuses everything until you resume, or revoke it, which is permanent. A revoked agent keeps its history and can never spend again.

Getting good behaviour

Three things make an agent predictable:

  • Name the agent in your prompt. Otherwise it guesses, and it may guess wrong.
  • State a ceiling. It cannot exceed its real limits either way, but a stated number gives it something to plan against.
  • Say what to do when refused. "Tell me what you can afford instead" turns a dead end into a useful answer.

The Capabilities screen has prompts already filled in with your own agent's details, ready to copy.

On this page