An agent with no configuration is a stranger who shows up every morning having forgotten the job. Configuration is the file that hands it the job description before it starts. You already wrote your project's rules; this chapter defines a specific agent with a role of its own.
3.5.1Give each agent a role
A role is a small file, not a paragraph you retype into chat each time. In Claude Code it lives in .claude/agents/, and every serious tool has the same idea: a YAML frontmatter block plus instructions.
---
name: reviewer
description: Reviews code after it's written. Use before any commit.
tools: Read, Grep, Glob
---
You review for correctness, security, and style. You do not
write features and you do not refactor. Report issues by
severity, then stop. A hardcoded secret is a blocking issue.
That is a coworker defined in a file your project keeps, not a mood you set and lose each session.
3.5.2Skills and rules per agent
A role carries two kinds of attachment, and you keep them as separate files so they compose. A skill is a procedure loaded only when the task needs it. A rule is a hard line, always on.
.claude/
agents/reviewer.md (the role)
skills/security-review.md (loaded when reviewing auth code)
rules/no-secrets.md (always on, every turn)
That layout is Claude Code's; every serious tool has an equivalent folder. You never pour all of it into one giant prompt. You attach the right skill to the right role and let the rest stay out of the way until it is needed.
3.5.3Configuration keeps output consistent
Unconfigured agents drift: the same request gives a different shape of answer on Monday and Friday. A rule file pins the shape:
# rules/output.md
- Answer in bullets, never a wall of text.
- Back every claim about the system with a command you ran.
- No "should be" or "probably". Check, then state.
Now the quality bar lives in a file, not in your memory of how you happened to ask last time.
3.5.4Write it once, reuse everywhere
Configuration is an asset, not a chore. The role you define today serves every task that agent ever runs, and the rule you write once fires on every future turn.
An hour spent standing an agent up correctly pays back on every job it does after. You are not configuring a task, you are configuring an employee, and the file is committed alongside the code it governs.
This prompt has your agent write its own role file, in your tool's format:
Act as a senior engineer configuring my agents. Create a
role file for a reviewer agent, in the format my tool uses:
frontmatter with name, description, and the minimum tools it
needs, then standing instructions to review for correctness,
security, and style and then stop, treating a hardcoded
secret as a blocking issue. Keep it short.
My tool and project:
Do this now: paste the prompt so your agent writes the reviewer role file, then read it and tighten one line yourself.
