recursive-router

Route delegated audit and implementation work through external CLIs with verification.

128|10|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/try-works/recursive-mode --skill recursive-router
Or copy as Structured Prompt for Agentβ–Ό
Please help me install this Agent Skill.
Skill: recursive-router
Source: https://github.com/try-works/recursive-mode/tree/main/skills/recursive-router
Command: npx skills add https://github.com/try-works/recursive-mode --skill recursive-router

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Routed delegation is error-prone without auditable, policy-driven routing between CLIs/models; this skill provides a safe, policy-governed way to route delegated work while preserving controller verification and fallback behavior.

Core Features & Use Cases

  • Policy-driven routing discovery
  • Safe delegation to external CLIs/models with canonical context bundles
  • Audit-friendly run evidence and action records
  • Fallback and repair loops when routed outputs fail verification

Quick Start

Configure a routing policy for this repository to delegate tasks to an external CLI or model.

Frequently Asked Questions about recursive-router

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

FAQPage Schema
How do I route delegated work to external CLIs safely with verification?β–Ό

Safe routing of delegated work to external CLIs requires a canonical routing policy, discovery inventory, pre-dispatch context bundles, and an action-record system that enforces verification before acceptance.

What is policy-driven routing for delegated audit and review tasks?β–Ό

Policy-driven routing is a method to delegate audit or review tasks to external models using explicit context bundles, preserving controller verification and generating audit-friendly run evidence.

How do I set up auditable delegation to external models for a repository?β–Ό

To set up auditable delegation, configure a routing policy for your repository that defines how tasks are delegated to an external CLI or model, ensuring all routed outputs generate action records.

Does safe delegated routing support fallback and repair loops when verification fails?β–Ό

Safe delegated routing supports fallback and repair loops, automatically handling scenarios where routed outputs from external CLIs fail verification before any acceptance occurs.

Can I use bounded delegation for implementation work across local and remote tooling?β–Ό

You can use bounded delegation for implementation work across local and remote tooling by applying a canonical routing policy that enforces explicit verification and audit trails.

What do I need to implement policy-governed routing between CLIs and models?β–Ό

Implementing policy-governed routing between CLIs and models requires a discovery inventory, pre-dispatch context bundles, and an auditable action-record system to ensure safe delegation and fallback behavior.