2.1 KiB
detectors
The AIMsight anomaly detectors, and their improvement. One detector per anomaly family of the displacements, classified into events by scripts/categories.py; scripts/data.py reads a station folder. Standard library only (Python 3).
python3 scripts/categories.py <station folder> --from YYYY-MM-DD --to YYYY-MM-DD
The station folder: displacement/, metrics/, camera/ (the format of data/<STATION>/ in Diagnostic-app).
scripts/: the detectors (visibility.py,obstruction.py,target_issue.py,thermal_displacement.py,no_data.py,displacement.py),categories.py,data.py.scripts/wf-improve/: the wf-improve workflow's tools, run from the working repository:regression.py(the benchmark, all labels oflabeling/auto/),score.py(the score of a range),sensors.py(the sensors of each event).thresholds.md: the registry of the detectors' thresholds.
The wf-improve workflow: a Claude Code plugin
The repository is also a Claude Code plugin (.claude-plugin/plugin.json), to be mounted as detectors/ in a working repository that has labeling/ (the labels, labeling/auto/<STATION>.jsonl; the analyses, labeling/auto/analyses/), data/ and agent-tools/: Diagnostic-app, where it is a submodule.
skills/wf-improve/SKILL.md:/detectors:wf-improve <analysis>, the session that improves the detectors from an analysis of the Data page, without overfitting.agents/:upgrader(diagnoses and fixes the detectors) andvalidator(LLM-as-judge: keeps, removes or cancels the fixes), calleddetectors:<name>.wf-improve/: the sessions, one folder<session>_<analysis>/each, written by the skill.
From the working repository:
claude --plugin-dir detectors # then /detectors:wf-improve <analysis>
python3 detectors/scripts/wf-improve/regression.py # the benchmark
The session commits the detectors here, on main, without pushing: push detectors so that everyone has the updated detectors.
Copied from Support-Agent on 2026-11-13 (commit 48a3092); the detectors now evolve here.