0xArx

Build applications with the Claude Managed Agents API — Anthropic's managed infrastructure for running Claude as an autonomous agent in a cloud container. Trigger this skill whenever the user wants to build with managed agents, asks about agent sessions, wants to stream agent events, uses client.beta.agents / client.beta.sessions / client.beta.environments, asks about the managed-agents-2026-04-01 beta, wants Claude to autonomously run tools in a sandbox, or asks about outcomes, multiagent orchestration, or persistent agent memory. Also trigger when the user asks how to run Claude for long-running tasks, async work, or when they want Claude to execute bash commands, write files, or browse the web autonomously.

Design a tight, self-verifying autonomous LOOP for a coding task — grounded in THIS repo and THIS agent's real capabilities. Unlike a copy-paste loop generator, this skill first inventories the actual environment (installed skills, subagents, MCP servers/plugins, hooks, scheduled tasks, the codebase, git worktrees), then interviews you only for the gaps, wires the loop to assets that really exist, and can launch it. TRIGGER when the user says "loop coding", "design a loop", "start a loop", "loopcode this", "set up a loop for X", or asks to build an autonomous agent loop. SKIP for a one-shot edit the user just wants done now.