Subagents

Subagents let fx delegate independent work to another agent. Each has its own conversation, works in the same workspace, and reports back to the main agent.

Ask fx to delegate work that can run independently:

Have a subagent review the parser tests and report missing cases.
It should not edit files. Meanwhile, investigate the parse error.

One-off and named subagents

Create a named subagent, then continue it by name:

Create a subagent named reviewer to check the parser tests.
Ask reviewer whether the new test covers the missing case.

Named subagents belong to the current fx session and remain available when you resume it. Their conversations do not appear separately in fx resume.

While a subagent works

fx waits for the result by default. Send a follow-up prompt to keep working with the main agent while the subagent continues. Its result returns to the same conversation when it finishes.

You can also send feedback to a named subagent:

Tell reviewer to include invalid UTF-8 input in its review.

If the subagent is running a tool, fx applies the feedback after that tool finishes. Changing its saved instructions waits until the subagent is idle.

Choose a model

You can choose a different model and reasoning effort for each subagent. This lets you use a stronger model to plan and review, then delegate implementation to a lower-cost model.

For example, keep GPT-6 Astra in the main conversation and ask:

Have a subagent use Kimi K3 to implement this plan.
Return the changes and test results for review.

Subagents use the current model and reasoning effort unless you choose others when creating them. A named subagent keeps that selection for later prompts. See Models.

Access

Subagents work under the parent session's permission limits and cannot create other subagents. If several subagents edit files, give each one a separate area to avoid conflicting changes.