Specialist Agent
A specialist agent is an AI agent scoped to a specific role — coding, research, content, browsing — with role-shaped prompts, tools, and evaluation criteria.
In plain English
A specialist agent is a purpose-built agent with a narrow job. Instead of one generalist that must know how to do everything, you define a coder, a writer, a researcher, a browser, and give each a system prompt that reads more like a job description than a catch-all instruction set. Each specialist gets exactly the tools it needs — file edits for the coder, a web browser for the researcher — and nothing else.
The effect is dramatic. A specialist with a focused prompt and a three-tool belt makes fewer mistakes than a generalist with everything piled into one prompt. It also makes the system easier to reason about: you can audit each specialist in isolation, swap it out, or upgrade its model without touching the rest. Specialists are the unit of composition in a multi-agent system.
Why it matters for Black Box
Black Box ships 18 specialist agents — covering coding, content, research, browser, business ops, design, product, marketing, sales, social, SEO, ads, PR, customer success, support, data, finance, and legal — each with its own system prompt, toolset, and constitution. When the CEO agent delegates a task, it picks the specialist whose role matches. Owners can add custom specialists through Skill Packs without touching the core.
Examples
- A Content specialist with a brand-voice prompt and publishing tools, but no code-execution access.
- A Research specialist with a browser and a summarizer, but no ability to send email.
- A Coding specialist that can read, write, and run files inside a sandboxed workspace.