llms.sgit.ai / documents / wider-work
The Wider LLM Work
The estate's LLM material measured rather than described: provenance and determinism as the oldest thread, running back to an OWASP AppSec Lisbon talk in June 2024; grounding rather than prompting as the answer to fabrication; OpenRouter as the actual provider layer; local and offline models; prompt injection; and a table of what is thin, which includes no evals, no structured-output guidance and one file mentioning model routing.
What it produced on this site
The document, in full
Rendered from the raw file below. The raw markdown is the source of truth: this page is presentation, and nothing here is edited.