Add Citations and TL;DR

[balteravishay]

Add Citations and TL;DR sections to Security-Focused Guide for AI Code Assistant Instructions

[david-a-wheeler]

Catherine Tony et al's work is especially appropriate to this work: https://arxiv.org/abs/2407.07064v2

The work found that RCI was the most effective, and if you couldn't do that, detailed prompting (as described here) was next-best. From a scientific view I understand why they compared approaches in isolation to each other.

Yet from a practioner's view, I don't want to know "what one technique should I use". Nothing prevents us from using multiple techniques. I want to know "what combination of techniques would produce the best answers within various resource constraints"? The top two techniques were RCI (asking it to review its answer and correct it), followed by specific guidance on the results. Once my proposed addition of RCI is added to this text, we're incorporating both of the top 2 techniques. I hope there will continue to be experimentation. However, given the evidence I can currently find, that seems most likely to produce the best results. If someone finds a better way, we can update to match.

[david-a-wheeler]

Thanks so much! I contributed a lot of proposed tweaks which you've accepted. Copilot found a few formatting problems, which look legit. Could you please review them, and if they make sense, accept them?

Other than that, I think this new version is much better. Let's merge it & get WGs to review it. I think everyone will be delighted.

[david-a-wheeler]

I'm sorry that you had to suffer through an endless number of suggestions :-). But I'd like to think that the result now is lots better, and I didn't want to create merge conflicts.

Please merge at your convenience. This is great stuff.

[Copilot]

Pull Request Overview

This PR adds comprehensive citations and a TL;DR section to the Security-Focused Guide for AI Code Assistant Instructions. It transforms the guide from a basic instructional document into a well-researched, academically-backed resource with proper attribution and quick reference material.

Key Changes

• Added a TL;DR section with essential security principles and sample instructions for immediate use
• Incorporated extensive citations throughout the document to support recommendations with research evidence
• Added a comprehensive appendix with detailed citations and references section
• Updated navigation links across documentation to reference the new guide

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
docs/index.md	Added link to the Security-Focused Guide in the ongoing work examples
docs/developers.md	Added the guide as a resource for developers with descriptive text
docs/Security-Focused-Guide-for-AI-Code-Assistant-Instructions.md	Major expansion with TL;DR section, citations throughout, and references appendix