release-2.2

What's Changed

• Add flair to readme by @inverse in #77 -> The README now looks a bit more pretty and some typos were fixed.
• Updated README with fixed typos and info blocks by @paulotruta in #78 -> A little revamp of the README with lots of new info and warning blocks and more detailed explanation about all the commands available as part of the last release.
• Add support for llm_system_prompt in config.json by @paulotruta in #79 -> Now you can have custom llm prompts for your skill without having to modify a single line of code. If you use make commands to automatically manage your Alexa-Hosted skills, you can use different prompts per target too ;)

We hope you enjoy the new release! Please let us know if something does not work, or how this project is benefiting you! We'd love to hear your stories.

Full Changelog: release-2.1...release-2.2

release-2.1

What's Changed

This update brings a lot of fixes and improvements, introduces testing foundations, and allows you to automatically create and manage multiple Alexa Hosted Skills with different configuration files and invocation names, all via the command line with simple to use make new, make list, make config, and make update commands!

• Add testing foundations by @inverse in #71
• Fix typos in prompt by @inverse in #74
• Ensure we're running on Python 3.8 by @inverse in #75
• Extract client by @inverse in #72
• Repackage for CLI use + add debug / deploy support by @paulotruta in #73

Full Changelog: release-2.0...release-2.1

2.0

• This is a release that attends all major open issues after 1.0 version (alpha-1.0) stopped working completely.
• The code had a significant refactor and now supports calling LLM API's directly, as well as proxy the request into a generic webhook, together with a bunch of context data. I will be creating a blog post on https://wordpress.jpt.land going into more detail about how I personally use the webhook solution to power my Alexa skill response via a n8n automation workflow which itself uses the OpenAI API with a full fledged Agent framework (Langchain).
• We also added a nice little prompt for the LLM API's, which now can be tweaked to your needs.
• We personally tested the updated code to work and results are very pleasant, as this version is immensely superior to the previous release.
• We made this repository be a template, so it is easier than ever to start your own skill repository!
• We will work to bring a bunch of new capabilities to this skill! Starting with unit testing the code, much needed deployment automations, and improved documentation.
• We hope you find this useful! If you have any questions or would like to contribute, start an Issue!

What's Changed

• Minor improvements by @inverse in #1
• Create dependabot.yml by @inverse in #2
• Fix venv + remove config by @inverse in #9
• Updated interaction model and manifest, renamed HelloWorld intent to QuestionIntent by @paulotruta in #10
• Bump ask-sdk-core from 1.11.0 to 1.17.1 in /lambda by @dependabot in #5
• Bump boto3 from 1.9.216 to 1.26.27 in /lambda by @dependabot in #15
• Bump revchatgpt from 0.0.31.5 to 0.0.33.5 in /lambda by @dependabot in #14
• Tidy code and use pre-commit by @inverse in #8
• Add test script and tidy import by @inverse in #17
• Update dependabot.yml by @inverse in #22
• Bump actions/checkout from 3 to 4 by @dependabot in #63
• Bump actions/setup-python from 3 to 5 by @dependabot in #64
• Bump pre-commit/action from 3.0.0 to 3.0.1 by @dependabot in #65
• Updated code (Uses OpenAPI or Webhook and works more reliably) by @paulotruta in #62
• Update pre-commit to ruff by @inverse in #69
• Tidy init by @inverse in #68
• Canned handler responses by @paulotruta in #70

New Contributors

• @inverse made their first contribution in #1
• @paulotruta made their first contribution in #10
• @dependabot made their first contribution in #5

Full Changelog: alpha-01...release-2.0

1.0

Full Changelog: https://github.com/paulotruta/alexa-skill-llm-intent/commits/alpha-01