Skip to content

HELIOS_AI_SYSTEM.md

Latest commit

 

History

History
52 lines (34 loc) · 1.3 KB

HELIOS_AI_SYSTEM.md

File metadata and controls

52 lines (34 loc) · 1.3 KB

Helios AI Knowledge System

Purpose

Ask Helios should answer from the actual build, not from generic marketing copy.

Source priority

The AI system should prioritize:

  1. mirrored handoff documents
  2. operator and runbook documents
  3. review rebuttal and coverage docs
  4. selected build files and API routes
  5. model synthesis only after source context is loaded

Current answer sources

The current build-aware AI system is designed to use:

  • START.md
  • white paper and operator docs
  • tokenomics guide
  • review rebuttal docs
  • launch blueprint and runbook
  • selected build files such as app.py, api/routes.py, config.py, and core/handoff.py
  • GitHub repository references

Operator rule

Do not store live model credentials in code.

Set them through environment variables only.

Recommended environment

  • HELIOS_AI_API_KEY
  • HELIOS_AI_MODEL

What a good answer looks like

A good Ask Helios answer should:

  • answer the question directly
  • stay aligned to the current build
  • separate what exists from what is pending
  • reference source docs or code paths when possible
  • avoid hype and unsupported certainty

Final position

The AI system is part of the operator and stakeholder support layer. Its job is not to invent a story. Its job is to explain the current Helios build accurately.