refusal-routing

Route tasks to fallback models when the primary refuses, times out, or errors.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/0x-wzw/necroswarm --skill refusal-routing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: refusal-routing
Source: https://github.com/0x-wzw/necroswarm/tree/main/skills/refusal-routing
Command: npx skills add https://github.com/0x-wzw/necroswarm --skill refusal-routing

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automatically route to the next model in the fallback chain when a model refuses, times out, or errors, ensuring resilient task completion with minimal user disruption.

Core Features & Use Cases

  • Dimension-aware fallback: route within NecroSwarm dimensions to equivalent models.
  • Multi-chain fallbacks: supports multiple predefined task-type chains (complex, moderate, etc.).
  • Silent recovery: returns a next-step model or a final verdict with clear guidance for partial results if exhausted.

Quick Start

Provide a failed-model and error details to obtain the next fallback suggestion.

Frequently Asked Questions about refusal-routing

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

FAQPage Schema
How do I automatically route tasks to a fallback model when the primary model refuses?

To automatically route tasks to a fallback model when a primary model refuses, you provide the failed model and error details to obtain the next fallback suggestion. This enables silent recovery by switching to backup models.

What types of model failures trigger an automatic fallback chain?

Model failures that trigger an automatic fallback chain include refusals, timeouts, empty responses, errors, and dead models. The system detects these specific failure states to select the proper next model in the chain.

Can I set a maximum fallback depth for multi-chain model routing?

Yes, you can enforce a maximum fallback depth for multi-chain model routing. This limits the number of fallback attempts and produces a clear verdict with partial results guidance if the fallback chain is exhausted.

How does dimension-aware fallback work for equivalent models?

Dimension-aware fallback routes tasks within NecroSwarm dimensions to equivalent models. This ensures that when a primary model fails, the fallback logic selects a replacement model matching the same specific task dimensions.

Does model routing support multiple predefined task-type chains?

Yes, model routing supports multiple predefined task-type chains such as complex and moderate chains. This allows you to apply different fallback strategies tailored to the specific complexity of the task being routed.

What happens when all fallback models in the chain are exhausted?

When all fallback models in the chain are exhausted, the routing mechanism stops and returns a final verdict. This verdict includes clear guidance for handling any partial results obtained before the maximum fallback depth was reached.