Discovers agent persona files and dispatches selected agents in parallel for a task.
npx skills add https://github.com/Femad-6/my-skills --skill team-builder-femad-6
Please help me install this Agent Skill. Skill: team-builder Source: https://github.com/Femad-6/my-skills/tree/main/.github/skills/team-builder Command: npx skills add https://github.com/Femad-6/my-skills --skill team-builder-femad-6
High-intent search queries and answers about installing and using this skill.
Select up to five agent personas and the skill spawns them simultaneously using parallel Agent tool calls, each receiving the agent file content plus your task description. Results are collected and merged into a single synthesized report.
Place markdown agent files in ./agents or ~/.claude/agents using either domain subdirectories or a flat layout with shared filename prefixes. The first markdown heading becomes the agent name and the first paragraph becomes its description.
Agent files must be markdown containing a persona prompt with identity, rules, workflow, and deliverables. The first # Heading is used as the agent name; if missing, the name is derived from the filename.
Discovery fails when no markdown files exist in the probed paths: ./agents and ~/.claude/agents at both depths. The skill reports which paths were checked so you can place agent files in one of those directories.
Yes, a maximum of 5 agents per team is enforced at selection time. If you select more, the skill lists them alphabetically and asks you to narrow down or keep the first five.
Use parallel Agent calls when agents work independently on the same task without needing to communicate. TeamCreate is only appropriate when agents must debate or respond to each other's outputs.
idiaz01Discover agents from directories, present domain menus, and spawn selected agents in parallel.
SZolothReviews local Git changes with parallel specialized agents for prioritized, actionable findings.
HomericIntelligenceAudit repository structure, code quality, testing, security, and CI/CD with evidence-based grading.
Kaleb-RupeRuns multi-agent debates with visible transcripts where specialized agents respond to each other.
