Triage subagent requests and generate bounded prompts for specialized agents.
npx skills add https://github.com/dain-agency/mood-tracker --skill spawn-dain-agency
Please help me install this Agent Skill. Skill: spawn Source: https://github.com/dain-agency/mood-tracker/tree/main/.claude/skills/spawn Command: npx skills add https://github.com/dain-agency/mood-tracker --skill spawn-dain-agency
This Skill eliminates the common waste and low-quality output caused by poorly scoped subagent dispatches, which often produce placeholder code, skip project-specific patterns, and waste valuable context cache windows.
Invoke the spawn skill to triage your subagent request, select the appropriate specialised agent, and generate a bounded prompt for high-quality, scoped output.
High-intent search queries and answers about installing and using this skill.
Poorly scoped subagent dispatches cause placeholder code because they lack explicit context boundaries. Applying deliberate scope triage and bounded prompt generation aligns subagent outputs with your project patterns and user intent.
Generate bounded prompts for specialised agents by evaluating task complexity, selecting a project-specific agent, and creating a self-contained prompt with explicit goals, context, scope boundaries, and deliverables to prevent ambiguous outputs.
Subagent scope triage is the process of evaluating task complexity like file count and domain span to recommend the right number and type of subagents. You need it for non-trivial tasks including code implementation, review, research, and pipeline scaffolding.
Yes, workflow automation through scope triage supports multi-file feature scaffolding. It scores the scope by file count and domain span to recommend specialised agents, ensuring generated prompts enforce your component patterns across multiple files.
The best way to stop wasting context cache windows is to apply a scope scoring rubric before dispatching agents. This evaluates task complexity to recommend the right subagents and generates bounded prompts that prevent off-target outputs.
mattjeffersonConvert fuzzy requests into initiative specifications with measurable success criteria.
miles990Dispatch fresh subagents to execute multi-step implementation plans with two-stage reviews.
nsheapsOrchestrate multiple Claude Code sessions for parallel agent tasks.
Kgan01Score prompt quality and detect anti-patterns for Claude Code.
