ov.reporting.overview

The overview one-liner: observe -> analyze -> report -> synopsis (§4).

This is the pit-of-success entry point behind ov.overview(). It threads a single store through capture, the deterministic analyzers, report rendering, and the synopsis, returning the synopsis path/key. No model is involved.

ov.reporting.overview.run_overview(url: str, *, headed: bool = False, mode: str = 'reconstruct', probes: Any = 'default', store: Any = None, out_dir: Any = None, authorized: bool | None = None, lenses: tuple[str, ...] = ('ux', 'arch'), baseline: Any = None, **observe_kw: Any) → str

Run the full deterministic pipeline and return the synopsis path/key.

In mode="review" it inserts an own-target diff step between analysis and reporting: the run is compared against a stored prior baseline (baseline, or the latest prior run of the same target) so the review report and synopsis carry regression/drift detail. A first review run (no baseline) is a no-op.