The Ollama dashboard assessment was built entirely from pre-computed
statistical scores/reasons/counts — a deliberately tested compaction step
stripped all raw message content before it reached the model, presumably to
keep the prompt small for a local, GPU-constrained instance. That raw
content (sample_events/timeline, already truncated to short messages)
already flows through the pipeline; it was just filtered out at the last
step. Now the top-scoring field deviations and incidents each carry a
small, bounded number of real message excerpts, and the prompt explicitly
asks the model to flag suspicious content (injection attempts, unusual
command lines) that pure statistical scoring can't see — purely a
qualitative addition to the narrative note, never feeding back into scoring
or suppression.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Field roles are now inferred from observed values instead of names,
with src/dst disambiguation, value-pattern detector hints, an Ollama
advisor scoped to ambiguous fields only, confidence-gated auto-apply
with manual-field protection, and drift flagging for review
