agent-teamlisted

Auto-spawn a coordinated team of Claude Code teammates (each with own context, can DM each other, share a task list) when the user describes work that spans multiple domains and would benefit from specialists working in parallel with feedback loops. Triggers on phrases like "create a website for X", "build me an app for Y", "spin up a landing page for Z", "make me a site about W", "I need this built end-to-end", "review this from multiple angles", "debug this with competing hypotheses", or any request that implies multi-specialist parallel work with a quality loop. Auto-classifies the request, picks team composition + model-per-role + file ownership + hand-off graph, then spawns. Skip for single-domain tasks (use matching specialist skill), sequential pipelines (use subagents), or simple one-shot work. Requires the host's multi-agent team feature to be enabled; on Claude Code that means setting CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 in settings.json.
donatassimkus/claude-ai-skills · ★ 1 · AI & Automation · score 77

Install: claude install-skill donatassimkus/claude-ai-skills

# Agent Team Auto-orchestrates a parallel team of Claude Code teammates against a user goal. The user states what they want; this skill picks the team, the model per role, file ownership, hand-off graph, and the spawn prompt — without the user having to remember any of that. **Platform note.** The mechanics below are written against Claude Code's agent-teams feature, which is what the paths, the spawn phrasing, the plan-approval switch and the limitations section describe. The METHOD is not tied to it: classifying the request into a team shape, sizing the team, assigning file ownership, drawing the hand-off graph, tiering models by role, monitoring and redirecting early, and shutting down cleanly all transfer to any system that can run several agents in parallel with messages passing between them. On a different one, substitute its own spawn syntax, its config location, and its equivalent of plan approval, and treat the limitations section as a list of questions to answer about that platform rather than as facts about it. ## When to invoke User describes work that meets ANY of these: - **Multi-domain build:** "create / build / spin up / make me [a website / app / landing page / portal / dashboard / tool] for [topic / person / audience]." Spans front-end + back-end + content + QA. - **Quality loop:** "write me [content] with reviews," "build this with QA," "I want [thing] reviewed from multiple angles." Writer + reviewer(s) + verifier. - **Competing-hypothesis debug:** "th