domains / families / research / do-agent

do-agent

separate

This skill should be used when the user asks to "/do-agent", asks for "Execute complex tasks using multi-agent architecture with context protection", or needs the workflow previously provided by the /do-agent slash command.

When to use

• Execute complex tasks using multi-agent architecture
• Multi-stage execution plan with up to 10 subagents in parallel
• Systematic workflow: explore, ultrathink, plan, track, execute, review, revise, deliver
• Complete automation from planning to execution to review and revision

When not to use

• Tasks requiring human intervention during the process
• Tasks requiring plan approval before execution

Skill page

Source