LEX AI Missing Knowledge Roadmap

What the public LEX AI documentation still lacks, which questions have since been settled, and how to close a gap without reintroducing stale facts.

A
Written By Atakan KayaLast updated 3 months ago

LEX AI Missing Knowledge Roadmap

This page tracks what the public-facing LEX AI documentation still lacks. It is working material for the next Help Center expansion, not a product commitment.

Closed since the last revision

  • The setup command spelling — settled. It is lex setup-with-ai, hyphenated. lex setup_with_ai is a Python module path and fails as a command, so any page still showing the underscore form is a defect rather than a variant.
  • The mode roster — settled at ten modes: brief, forward, backward, edit, review, test, input, deploy, mvp_generator, mvp_completion. Any page still saying six is describing an older release.

Open gaps

  1. End-user screenshots and exact walkthroughs for each of the seven supported coding environments.
  2. Public-safe escalation guide, including which context to attach to a support request.
  3. Glossary for MCP, server, mode, tool surface, and workflow terminology.
  4. Public Git/GitHub behavior matrix with security-reviewed wording.
  5. Full dashboard field reference and safe action guidance, panel by panel.
  6. Environment-specific mode-switch troubleshooting playbook, including the tool-list refresh gesture each host needs.
  7. Complete run examples per mode with artifact samples, most urgently for the newer test, input, and deploy modes.
  8. Version and update policy, and what compatibility to expect across lex ai-update migrations.
  9. Product walkthrough screenshots and videos.
  10. Security and privacy page covering local execution, remote calls, credential masking, and data handling.
  11. Guidance on model choice that points readers at lex ai-faq, rather than reproducing a ranking that changes as models are re-scored.

How to use this list

Take one item, verify it against the installed package rather than against an older article, and write it as a topic page. The corpus stays accurate when each fact lives in exactly one place and the other pages cross-reference it. Most of the stale claims found in the last revision existed because a number had been restated in four articles and corrected in one.

Was this helpful?

Your feedback shapes what we write next.