Practical overview
When it fits
Use it when you want a customizable Alexa skill template to connect a voice interaction model with a compatible LLM API, rather than a standalone chatbot or model service.
Who it is for
- Developers building an Alexa voice interface for an LLM.
- Alexa skill authors who want a configurable starting point for conversational question answering.
- Developers adapting a voice skill to an OpenAI-schema-compatible provider.
Documented uses
- Create an Alexa skill that accepts spoken questions and delivers LLM answers aloud.
- Configure a provider endpoint, model and API key for the skill.
- Set up or update an Alexa skill through the documented Makefile, Developer Console or ASK CLI paths.
Requirements & limits
Using the template requires an Alexa Developer account. The README also lists an ASK CLI for its Makefile workflow; Python 3.8 is optional for local development and an AWS account is optional for advanced deployment.
- The README names OpenAI, Anthropic and OpenRouter as compatible provider examples.
- The Makefile wizard is intended for Interaction Model skills written in Python and hosted by Alexa.
- Manual setup and additional locales require setting the invocation name in the corresponding locale interaction-model file.
- Hosted-skill updates run from the local repository to the Alexa Developer Console and overwrite console-side changes.
- The README describes the debug command as not fully tested and identifies a regional limitation for debugging Alexa Hosted skills.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Configure the LLM and invocation name
Prepare the template configuration values required for a skill to call a compatible LLM provider.
Before you start
- An Alexa Developer account
- An OpenAI API-schema-compatible provider URL, API key and model name
Steps
Copy config.example.json to config.json.
Set invocation_name to the words users will say to open the skill.
Set llm_url, llm_model and llm_key to the chosen provider endpoint, model and API key.
For manual setup or support for multiple locales, set invocationName in the relevant skill-package/interactionModels/custom/.json file.
Expected outcome
A configuration file contains the skill invocation name and provider connection values; locale interaction models can carry their own invocation name.
Guide limitations
- Automatic invocation-name configuration applies only to the Makefile deployment method and en-US locale.
- The guide covers configuration, not a verified deployment or successful provider connection.
Create an Alexa Hosted skill with the Makefile workflow
Start a new Alexa Hosted skill project using the template's documented Makefile wizard.
Before you start
- An Alexa Developer account
- ASK CLI installed and configured with the Amazon Developer account
Steps
Run make new in the repository.
In the wizard, choose Interaction Model, Python and Alexa Hosted for the modeling stack, programming language and backend hosting method.
Expected outcome
The wizard begins creating a new Alexa Hosted skill target using the template's intended options.
Guide limitations
- The Makefile workflow currently supports Alexa Hosted skills only.
- The README warns that choosing other wizard options could cause the process to fail.
- This setup guide does not include subsequent configuration or deployment steps.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 8 October 2026
- Original website title
- paulotruta/alexa-skill-llm-intent · Official homepage title · retained in its source language
Related options
An OpenAI-schema-compatible LLM provider — A model service used by the skill to generate answers; the template names OpenAI, Anthropic and OpenRouter as examples.