xiaohongshu-session-readerlisted

# Xiaohongshu Session Reader ## Scope - Reuse local Chrome logged-in session only. - Prefer HTTP/API extraction for stability and speed. - Use Codex Chrome plugin when HTTP/API path is blocked and visible logged-in browser state is needed. - Use Playwright only when explicitly requested/enabled and Chrome plugin cannot complete the fallback. - Support: - profile card titles (for 谁是卧底词组 collection) - note detail (title/desc/author) - comments when API permits; otherwise explicit fallback signal - Do not forge identity, bypass captcha, or brute-force anti-bot checks. ## Prerequisite Check Before extraction: ```bash python3 --version ``` Fallback-only prerequisite: ```bash codex mcp get playwright-ext # only for explicit Playwright fallback ``` ## Workflow 1. Run HTTP-first reader: ```bash python3 <skill_dir>/scripts/xhs_http_reader.py \ --url "<xhslink_or_xiaohongshu_url>" \ --mode auto \ --max-items 40 \ --max-comments 20 \ --pretty ``` 2. If response has `fallback.required=false`, use returned data directly. 3. If response has `fallback.required=true`, switch to Codex Chrome plugin fallback: - profile: open profile page and read visible card titles. - note comments: open note detail and read comments from DOM snapshot. - use Playwright only if the user explicitly asks for it and `playwright-ext` is enabled. 4. Normalize four-word groups with `scripts/undercover_parser.py`. 5. Persist progress in `.cache/xhs_undercover_progress.json` for ba