gf-router

Route user intents to GateFlow skills or agents with confidence scoring.

101|13|Updated Jan 28, 2026
One-click install
npx skills add https://github.com/codejunkie99/Gateflow-Plugin --skill gf-router
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gf-router
Source: https://github.com/codejunkie99/Gateflow-Plugin/tree/main/plugins/gateflow/skills/gf-router
Command: npx skills add https://github.com/codejunkie99/Gateflow-Plugin --skill gf-router

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill acts as the routing intelligence for GateFlow, semantically classifying user intent and directing requests to the appropriate skill or agent while managing ambiguity with expand mode.

Core Features & Use Cases

  • Semantic intent classification: Determine user goals based on meaning rather than keywords.
  • Confidence scoring and routing: Decide between direct handoff or expanded dialogue to resolve ambiguity.
  • Handoff orchestration: Build rich context and route to the correct target (gf or gateflow agents) with structured context.

Quick Start

Say: "I need to route this user request to the correct GateFlow skill." GateFlow will classify and hand off to the most suitable target while handling ambiguity through expand mode.

Frequently Asked Questions about gf-router

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

FAQPage Schema
How does semantic intent routing work for agent orchestration?

Semantic intent routing classifies user goals based on meaning rather than keywords, computing a confidence score to determine whether to directly hand off the request or trigger expanded dialogue to resolve ambiguity.

How do I route ambiguous user requests to the correct agent skill?

To route ambiguous user requests, the system computes a confidence score during intent classification; if ambiguity is detected, it triggers expand mode to gather clarification before orchestrating a structured handoff.

What is the best way to build rich context before handing off to another agent?

The best way to build rich context for handoffs is using a structured routing protocol that packages the classified intent and confidence score into a structured context object handed off to the target agent.

Does this intent classification approach use keyword matching or semantic understanding?

This intent routing approach uses semantic understanding rather than strict keyword matching, allowing it to determine user goals based on the contextual meaning of the request.

Can I use expand mode to resolve low confidence routing decisions?

Yes, expand mode is specifically triggered when the confidence score indicates ambiguity, enabling the system to resolve low confidence routing decisions through expanded dialogue before finalizing the handoff.

When should I not use automated intent routing for agent handoffs?

Automated intent routing may not suit contexts requiring deterministic command dispatch, as it relies on semantic classification and confidence scoring which introduces probabilistic handling and expand mode interactions for ambiguous requests.