Capturing paginated JSON from an authenticated research UI

What this skill carries — grep references/ for any subject the names below miss: !d=${CLAUDE_SKILL_DIR}; command -v skill-toc >/dev/null 2>&1 && exec skill-toc "$d"; s=$HOME/.claude/skills/plugin-utils/bin/skill-toc; [ -x "$s" ] && exec "$s" "$d"; echo "(skill-toc unavailable: references and scripts are NOT listed here — install the plugin-utils plugin, or start a new session so its bin/ reaches PATH)"

Drive the user's logged-in browser over CDP, hook XMLHttpRequest, click through the result pages, and keep the JSON the page already received. No direct API calls, no credential handling, no fighting the export cap.

Decide the access route BEFORE touching anything

Systematic capture of a paid research corpus is the user's decision to make knowingly, not yours to slide into. Surface the trade-off once, in plain terms, then do what they decide.