houmao-agent-loop-pairwise-v4

Generate and operate template-driven Houmao pairwise-v4 tree loop plans with run-control actions.

18|3|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/igamenovoer/houmao --skill houmao-agent-loop-pairwise-v4
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: houmao-agent-loop-pairwise-v4
Source: https://github.com/igamenovoer/houmao/tree/main/src/houmao/agents/assets/system_skills/legacy/houmao-agent-loop-pairwise-v4
Command: npx skills add https://github.com/igamenovoer/houmao --skill houmao-agent-loop-pairwise-v4

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Enables users to explicitly create and operate a legacy “pairwise-v4” Houmao agent loop that is template-driven, workspace-aware, and bound to a strict document template contract with coverage auditing.

Core Features & Use Cases

  • Template-driven plan authoring (plan / initialize): Generates a template-driven enriched tree loop plan written to a user-selected output directory, including workspace contract artifacts and a constraint coverage audit.
  • Deterministic run control (start / peek / ping / pause / resume / recover_and_continue / stop / hard-kill): Supports the canonical lifecycle actions for authoring, preparing, starting, operating, pausing, resuming, recovering, stopping, and emergency hard-killing.
  • Guardrails for strict workspace + constraint preservation: Records whether the run uses standard or custom workspace contract mode and enforces “fail-closed” requirements for mailbox and gateway capability for required participants.
  • Memo-first durable guidance & routing-packet validation: Performs routing-packet validation and writes run-owned durable memo guidance via the initialize flow.

Quick Start

Use the Skill only when you explicitly request houmao-agent-loop-pairwise-v4 and you have a plan output directory, then ask it to run plan or initialize to generate or validate the template-driven workspace-aware tree loop artifacts.

Frequently Asked Questions about houmao-agent-loop-pairwise-v4

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate CLI-based AI agents with strict workspace contracts and routing packets?

You can orchestrate CLI-based AI agents by authoring a template-driven enriched tree loop plan that generates workspace contracts, routing packets, and constraint coverage audits to manage agent teams.

How does memo management and recovery work in an AI agent loop?

Memo management writes durable run-owned guidance during initialization, while recovery uses the recover_and_continue lifecycle action to resume agent execution from failures within the tree loop.

Can I pause and resume a multi-agent tree loop plan during execution?

Yes, deterministic run control supports canonical lifecycle actions including pause, resume, ping, and stop to manage the execution state of the multi-agent tree loop plan.

What is a fail-closed verification requirement for agent mailbox and gateway capabilities?

Fail-closed verification enforces strict workspace preservation by requiring validated routing packets and confirmed mailbox or gateway capabilities for required participants before initialization.

When do I need a constraint coverage audit for AI agent orchestration?

You need a constraint coverage audit when generating a template-driven plan bundle to ensure strict preservation of source constraints across workspace contract artifacts and reporting templates.

How do I hard-kill a runaway AI agent loop process?

You can terminate a runaway agent loop process by invoking the hard-kill run-control action, which provides emergency lifecycle termination alongside standard stop operations.