Configure Subagents in Plain English Without Memorising API Parameters
When spawning subagents in Claude Code, you don't need to memorise any formal parameter names. Just describe what you want in plain English and Claude maps it to the right settings automatically.
For example, instead of trying to remember mode: "bypassPermissions" or run_in_background: true, you can just write:
Research X using a sonnet agent in background with bypass permissions
Claude understands natural language for all the key subagent options:
- Model: "use sonnet/opus/haiku for this"
- Mode: "run in bypass/plan/auto mode"
- Background: "run this in background"
- Isolation: "run in an isolated worktree"
- Role: "use a coder/researcher/tester agent"
You can mix these per-agent in a single instruction too: "Spawn 3 agents — the researcher in default mode, but the coder in bypass mode using Sonnet."
This works because Claude Code interprets your intent and sets the underlying parameters for you. You never have to remember the difference between bypassPermissions, acceptEditsToCreatedFiles, and plan mode names.
It also means you can be less precise and still get the right result. "Full auto on this one" will set bypass permissions. "Think through this carefully before touching anything" will set plan mode.
You've had this all along — just tell Claude what you want and it will wire up the right settings.
via @suhitanantula
Log in to leave a comment.
Before jumping to implementation, describe the problem conversationally and let Claude be your thinking partner. It asks clarifying questions, surfaces tradeoffs you haven't considered, suggests approaches, and pokes holes in your plan — so by the time you say "ok, build it," both of you know exactly what to build and why.
Paste error messages with "why did this fail?" instead of "fix this" to get Claude to diagnose the root cause before applying a fix.
When you need Claude to make changes in one area without affecting another, add negative constraints — "fix the bug but don't change the public API", "refactor the internals but don't create new files", or "update the logic but don't modify any tests." Explicit exclusions prevent Claude from making well-intentioned changes you'll have to undo.