perf-theory-tester

A modular runtime and orchestration system for AI agents, installable as an npm package, providing structured execution and coordination of AI workflows.

A modular runtime and orchestration system for AI agents, installable as an npm package, providing structured execution and coordination of AI workflows.

A modular runtime and orchestration system for AI agents that handles task selection, branch management, code review, CI, PR comments, and deployment through structured pipelines, gated phases, and persistent state across sessions. Includes 19 plugins, 49 agents, and 41 skills across multiple platforms including Claude Code, Codex CLI, and Cursor.

A skill from AgentSys, a modular runtime and orchestration system for AI agents with 26 plugins, 50 agents, and 47 skills across 5 platforms. Supports Claude Code, Codex CLI, OpenCode, Cursor, and Kiro with continuous updates.

A modular runtime and orchestration system for AI agents that manages structured pipelines, gated phases, and persistent state across session boundaries. Includes 26 plugins, 50 agents, and 47 skills for automating tasks like branch management, code review, CI, PR comments, and deployment across Claude Code, Codex CLI, and other platforms.

Part of AgentSys, a modular runtime and orchestration system for AI agents with 26 plugins, 50 agents, 47 skills, and 3,500+ tests, handling task selection, branch management, code review, CI, and deployment across 5 platforms.

A skill from AgentSys that validates delivery workflows in AI agent pipelines. AgentSys is a modular runtime and orchestration system with 26 plugins, 50 agents, and 47 skills across 5 platforms, built for Claude Code, Codex CLI, and other AI coding tools.

A modular runtime and orchestration system for AI agents with 26 plugins, 50 agents, 47 skills, and 3,500+ tests, built for Claude Code, Codex CLI, OpenCode, Cursor, and Kiro across 5 platforms.

A performance analysis skill within AgentSys that synthesizes performance investigation findings into actionable recommendations, part of a structured /perf pipeline alongside baseline, profiling, theory-gathering, and theory-testing skills.

Part of AgentSys, a modular runtime and orchestration system for AI agents that provides 26 plugins, 50 agents, and 47 skills across 5 platforms. Built for Claude Code, Codex CLI, OpenCode, Cursor, and Kiro, it serves as both a marketplace and installer for distributed agent capabilities.