/* THE COMPANY FILTER ON /member/{bioguide_id}. WHY THIS FILE EXISTS AT ALL. member_history.html shipped on 09/08 with six class names and this note attached to them: "the class names below are hooks for a later CSS-only commit, and until then this renders as a legible row of links rather than as a broken component". The later commit never happened. `.ticker-pill` has had no declaration anywhere in the nine stylesheets since, so the filter has been a bare row of underlined anchors - correct markup, passing tests, and not the control anyone designed. That is worth naming because it is the mirror image of the failure APP_ASSET_V exists to prevent: there, a rule is edited and never reaches the browser; here, markup is written and no rule is ever authored. Both are green in CI and both are only visible by opening the page. tests/test_ticker_filter_style.py now checks that every class the template uses has a rule here. SAME ARRANGEMENT AS lifecycle.css, signal.css, row-click.css AND stacked-names.css: one file, one feature, linked last in base.html, so "take the filter back to plain links" is deleting this file and one line rather than excavating rules out of a 34 KB brand layer. NOTHING IS INVENTED HERE. The pill is .q-chip's geometry from quorum.css - mono, 0.62rem, 0.08em tracking, 2px radius, one hairline border - and its active state is the .q-chip-2 / .band-strong treatment the rest of the site already uses for "this one", gold on a translucent gold plate. A seventh visual register for a filter would be a decision nobody asked for. THE SPECIFICITY TRAP DOES NOT APPLY HERE, and it is worth saying so because it applies almost everywhere else. style.css declares `table.leaderboard a` at 0-1-2, which beats any single-class rule - it is how the anonymous CTA button ended up Bone on Bone. This