Skip to content

external-ai-engineering-references.md

Latest commit

 

History

History
30 lines (22 loc) · 1.36 KB

external-ai-engineering-references.md

File metadata and controls

30 lines (22 loc) · 1.36 KB
Raw

External AI Engineering References

Repositories such as patchy631/ai-engineering-hub are useful implementation catalogs. Curated indexes such as owainlewis/awesome-artificial-intelligence are useful discovery sources. SET treats both as references only.

Boundary

  • no dependency, vendoring, submodule, installer, or runtime adoption;
  • no provider, MCP, vector DB, scheduler, or secret added from a reference alone;
  • no production claim until a local contract, test, and owner approval exist;
  • no bypass of proposal lifecycle, operation receipts, skill-quality validation, or data-egress review.

Reference Fields

Use this shape when a downstream repo catalogs examples:

category:
source project path:
possible local use:
required adaptation:
secrets/data-egress risk:
allowed pilot projects:
status: reference | candidate | piloted | rejected

Good first consumers are private/local AI repos such as CortexABV and CoqPi, where RAG, memory, voice/context, mock evaluation, and local fallback examples can be reviewed without changing production routes.

For curated awesome-list style sources, use docs/ai-engineering-reference-index-note.md before promoting any entry from reference to candidate.