acp-router

Routes user requests to appropriate ACP harness workflows via runtime or direct acpx paths.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/191854169/openclawFork --skill acp-router-191854169
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: acp-router
Source: https://github.com/191854169/openclawFork/tree/main/extensions/acpx/skills/acp-router
Command: npx skills add https://github.com/191854169/openclawFork --skill acp-router-191854169

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Route user intent to the appropriate ACP harness workflows (runtime-based or direct acpx) without resorting to subagent runtime, streamlining how tasks are executed across harness backends.

Core Features & Use Cases

  • Dual-path routing: selects between the OpenClaw ACP runtime path and the direct acpx path based on user intent and environment.
  • Thread-spawn governance: manages ACP harness session lifecycle in threads with explicit agentId mappings.
  • Recovery & resilience: implements recovery policies for ACP backends, including local repair prompts and restart options.
  • Fallback behavior: provides clear error messaging and fallback options when backends are unavailable or misconfigured.

Quick Start

Spawn an ACP harness session in a thread and run the requested code harness task.

Frequently Asked Questions about acp-router

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

FAQPage Schema
How do I route requests to ACP harnesses without using subagent runtime?

ACP routing directs user intent to appropriate harness workflows by selecting between the OpenClaw ACP runtime path using sessions_spawn and the direct acpx path, avoiding subagent runtime unless explicitly necessary.

How does thread-spawn recovery work for ACP harness sessions?

Thread-spawn recovery manages ACP harness session lifecycles within threads using explicit agentId mappings, applying local repair prompts and restart policies to restore unresponsive or misconfigured backends.

What happens when an ACP backend is unavailable or misconfigured?

When ACP backends are unavailable or misconfigured, the routing logic enforces clear fallback behavior by providing explicit error messaging and alternative fallback options to maintain task execution continuity.

Do I need explicit agentId mapping to spawn an ACP harness session?

Yes, explicit agentId mapping is strictly enforced when spawning ACP harness sessions in threads to ensure proper governance and routing of tasks across both runtime and direct acpx paths.

When should I choose the ACP runtime path over the direct acpx path?

Dual-path routing selects between the ACP runtime path and the direct acpx path based on user intent and environment, choosing the runtime path for session-managed workflows and acpx for direct execution.