Design your agent, download the kit.

No, and the difference matters. A kit is the complete specification of one agent: what it does, step by step, the tools it may call and what each one is allowed to change, the records it keeps, the rules that stop it acting beyond its authority, and the tests that prove it works. It does not connect to your systems by itself and contains no credentials.

You, or whoever builds for you, load it into the tools you already use: OpenAI or Anthropic models with tool calling, any MCP client, or an automation platform like n8n, Make or Zapier. Every file is plain JSON or Markdown, so nothing locks you to one vendor.

The agent's instructions, written for your business with your rules built in. Its tools, in OpenAI, Anthropic and MCP formats, each marked read, internal or external. A workflow per capability: what starts it, the steps in order, and which steps wait for approval. JSON Schemas for every record it keeps, including an approval queue and an audit log. An autonomy policy with four levels. A setup checklist of the facts only you can supply, integration notes for each of your systems, and acceptance tests.

Press "See your kit" above and the full file list for your design appears, before you pay anything. Want to read what is inside one first? Here is a complete sample kit, every file, free.

Because the size of the agent decides it, not a guess. The price is worked out from what you've actually designed: how many systems it has to operate, how many channels it works across, and how much judgment the job needs. The panel tells you which kit you're in and exactly why.

If it says Deputy and you wanted the Operator price, take something out and watch it flip back. A smaller agent genuinely is less to specify, and a Deputy key also covers an Operator-sized design if you trim it later.

Yes. One key is one agent: change the capabilities, the systems or your rules and download it again, as often as you need. A key is tied to the first job it downloads, so a different agent needs its own kit.

Your design is kept in this browser, so after paying you come back to it exactly as you left it. On another device, rebuild the same design and use the same key.

Everything on email, web chat, a shared inbox, Slack or Teams can run as soon as it is built. Inbound phone needs a number, not a registration.

The one genuine exception is SMS. Texting your customers from a US business number requires A2P 10DLC carrier registration, and part of it is a manual review nobody can hurry, so budget weeks rather than days. The kit flags it where it applies. Build and test everything else meanwhile, and switch SMS on when the registration clears.

The kit is written so that it starts at Draft, where anything a customer would see waits for a person. Every tool carries its effect, and the autonomy policy says which effects each level allows, so whoever builds it can enforce the rules in code rather than only in the prompt. Every action is logged before it happens, opt-outs are permanent, nothing is sent in quiet hours, and instructions that arrive inside a customer's message are treated as information, never as orders.

How it behaves once built is under your control, which is why the kit keeps it at Draft until its acceptance tests pass. How agents are controlled covers credentials, logs and the stop in full.

Look before you buy: the full file list and a summary of everything in the kit are on this page first. After buying, Gumroad's money-back guarantee shown on the product page applies. A refunded key stops unlocking downloads.

BlvkWare is William Russell Wheeler, in Mississippi. Kits are generated from one catalog of 23 roles and 73 capabilities, each with its own written acceptance test, the same catalog the agent runtime behind them was built and tested against.