Kits you install in an AI agent that works for you, such as Claude Code, Codex or Cursor, each with a guide to read.
Working with agents
- Working with agents
Procedure kit
A written procedure that a person and an AI agent carry out the same way. - Working with agents
Memory kit
Every correction you make is recorded, and the agent does not repeat the same mistake. - Working with agents
Verification kit
Before the agent says it is done, it checks that it really happened. - Working with agents
Orchestrator kit
One chat runs all the others: it hands out the work, and checks every result itself. - Working with agents
Reply kit
Every reply from the agent: a status line, the result, and nothing you didn't ask for. - Working with agents
Project kit
Your agent starts every project already knowing where it is. - Working with agents
Team kit
Your agent gets a team of specialist helpers, and hands each task to the helper that fits. - Working with agents
Command kit
The prompt you type almost every day, as a ready command your agent gets the same way every time. - Working with agents
Continuity kit
When a long chat is cut down to a summary, your agent does not retry what already failed. - Working with agents
Calm kit
Your agent tells you when it finishes, replies in the language you wrote in, and never sets off on a half-filled request.
Marketing and leads
Automation
Content
Web development
- Web development
Website kit
An AI agent that opens your site reads your content clean and correct, instead of guessing. - Web development
Headline kit
Your agent breaks every headline where the sentence breaks, not wherever the screen runs out. - Web development
Forms kit
Your agent builds your Hebrew form from components already tested, so it works right-to-left on the first try.
