Autonomous Agent for Claude Code
A Claude Code plugin that adds a self-learning autonomous agent system of 36 specialized subagents across four tiers, plus skills and slash commands for planning, implementing, reviewing and validating work automatically.
Autonomous Agent for Claude Code is a plugin that adds an autonomous agent system to Claude Code, organized into four collaborative groups: a strategic-analysis group of agents that analyze and recommend with confidence scores, a decision-making group that evaluates and creates execution plans, an execution group that implements changes, and a validation group that checks and provides feedback, for 36 specialized subagents in total including a master orchestrator.
It ships 27 skills and 41 slash commands across 10 categories, along with a real-time dashboard for monitoring active agents and their tasks, pattern learning intended to improve suggestions over repeated similar tasks, full-stack validation, and OWASP-based security scanning with automatic fixes for some findings. The project describes its processing as entirely local, with no data sent to an external service beyond the model provider the plugin is configured to use.
It works across models rather than being tied to one, and the project describes itself as privacy-first and cross-platform.