on-keyday/agent-harness
Parallel Claude Code harness — dispatch tasks to multiple agents across repos/hosts, attach interactively, and let agents message each other over a per-task broker. Toy/dogfood scope.
Parallel Claude Code harness — dispatch tasks to multiple agents across repos/hosts, attach interactively, and let agents message each other over a per-task broker. Toy/dogfood scope.
npx add-skill on-keyday/agent-harnessParallel Claude Code harness — dispatch tasks to multiple agents across repos/hosts, attach interactively, and let agents message each other over a per-task broker. Toy/dogfood scope.
Intelligent automation and multi-agent orchestration for Claude Code
Open-source coordination backend for AI agent teams. Shared memory, messaging, tasks via MCP. 20 tables, 75 tools. Convex-native.
A reproducible operating model for running team-led OpenClaw agents with Work Units, evidence, and alert-only monitoring.
Sync Claude Code and Codex skills, instructions and subagents across devices
Mandrel is an orchestration and harness for agentic AI coding. It ships a system prompt, personas, a two-tier skill library, baseline rules, slash-command workflows, and a GitHub-backed orchestration engine — distributed as a Git submodule into your project's .agents/ directory.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.