input-branching

Route voice inputs to processing paths based on content prefixes.

10|2|Updated Mar 24, 2026
One-click install
npx skills add https://github.com/autonomous-ai/autonomous-lamp --skill input-branching
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: input-branching
Source: https://github.com/autonomous-ai/autonomous-lamp/tree/main/lamp/resources/openclaw-skills/input-branching
Command: npx skills add https://github.com/autonomous-ai/autonomous-lamp --skill input-branching

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Understand and route voice inputs from the realtime voice agent pipeline to the appropriate processing paths based on content prefixes.

Core Features & Use Cases

  • Delegated with voice instruction: handle [voice-instruction] and [transcript] inputs to drive explicit actions.
  • Delegated without instruction (fallback): process inputs as-is when no instruction is provided.
  • Handled by realtime agent (chit-chat): respect [HANDLED]/[REPLY] interactions and update context without issuing new commands.

Quick Start

Provide a voice event with either [voice-instruction]/[transcript], or [HANDLED]/[REPLY], and the system will route it to the appropriate processing path.

Frequently Asked Questions about input-branching

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

FAQPage Schema
How do I route voice inputs from a realtime agent to different processing paths?

Route voice inputs by using tag-based content prefixes like [voice-instruction], [transcript], and [HANDLED]/[REPLY] to direct realtime agent events to the correct processing path for tool calls or memory updates.

What is the best way to handle voice transcripts when no explicit instruction is provided?

Handle voice transcripts without instructions by processing the delegated input as-is. The fallback path processes plain inputs directly when no explicit voice instruction tags are detected.

How does a realtime voice agent manage chit-chat without triggering new tool calls?

Realtime voice agents manage chit-chat by respecting [HANDLED]/[REPLY] tags to update context memory and generate responses without issuing new commands or triggering downstream tool calls.

Can I use tag-based prefixes to guide voice instruction routing and memory updates?

Yes, you can use tag-based prefixes like [voice-instruction] and [transcript] to guide voice instruction routing, explicitly driving tool calls, memory updates, and response generation based on content.

What formats does the voice input routing system support for delegated agent handoffs?

The voice input routing system supports delegated agent handoffs using tag-based formats including [voice-instruction] with summarized transcripts, plain delegated inputs, and [HANDLED]/[REPLY] interactions.

Why are my voice agent transcripts not triggering the correct tool calls?

Voice agent transcripts may fail to trigger tool calls if required content prefixes like [voice-instruction] or [transcript] are missing, causing inputs to fall into the plain delegated processing path.