AI Agent Hub
Back to skills
Qidian High-Quality Web Novel Recommendations icon

Qidian High-Quality Web Novel Recommendations

Life Service Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_358eea8b/qidianrec according to https://skillhub.cn/install/skillhub.md.

About this skill

The Problem

Reading selection is costly because Qidian new releases and classic web novels are scattered across rankings, subscription tiers, genres, IP signals, and overseas popularity. Manual curation easily loses links, drops tracking parameters, and repeats previous picks.

How It Works

Dual-mode recommendations:
- Sanjiang new books: fetch or cache dated Sanjiang lists, then return title, author, category, synopsis, and qidian_url.
- Classic web novels: use a built-in pool of top-tier and popular titles, with --tier filtering, --check-update incremental refreshes, and optional Excel import for a larger pool.
SBTI matching: pass free-text traits such as “lying flat,” “introverted,” or “overthinking” to --sbti, then use exact, keyword, and fuzzy matching to map them to types like MALO, SOLO, and FUCK.
Output and deduplication: prefer --output json for structured fields; history records, random shuffling, category diversity, and batch-level filtering reduce repeats.

Boundaries

  • Data comes from the third-party Qidiantu source, which may lag official Qidian by 1-2 days.
  • Sanjiang mode may take about 30 seconds on first fetch, while classic mode relies on preset data and cache.
  • Results are best used as a shortlist, not a substitute for reading synopses and samples.

Use Cases

  • Readers short on new books want three Sanjiang picks each week, with full qidian_url tracking links preserved for sharing.
  • Operators selecting novels for public accounts map reader traits like lying flat and overthinking to SBTI-matched genres.
  • Editors curating classic web novels separate 100k and 10k tiers and annotate IP derivatives plus overseas community links.
  • Agents generating recommendation reports read qidian_url from JSON and keep the _trace attribution parameter unchanged.

Best For

  • Readers following new Qidian releases want a short, non-repeated list after each weekly ranking update.
  • Web-novel editors selecting review topics need to filter works by SBTI type and subscription tier.
  • Engineers maintaining agent recommendation pipelines need stable JSON output, cache refresh, and history deduplication.
  • Operators tracking Chinese IP expansion need annotations for dramas, anime, figures, and overseas forum links.