story-long-scan: Chinese web-novel market rank scanning (Qidian, Fanqie, Jinjiang…)
Scrape platform rankings, aggregate genre/word-count/tag trends into an actionable market report — written in Chinese, for the Chinese web-novel market
- What
- Scrape platform rankings, aggregate genre/word-count/tag trends into an actionable market report — written in Chinese, for the Chinese web-novel market
- Cost
- Free
- Needs
- Node.js (the bundled ranking scrapers and the aggregation script run with node); a Chrome session via this repo's companion browser-cdp skill for the platforms and chart types that need a browser (Qidian's default charts do not); real ranking data, scraped by the scripts, pasted or supplied by you, or, as a last resort, the skill's built-in knowledge; the skill is written in Chinese
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @zenstory-ai's story-long-scan: a long-form web-novel market analyst for the Chinese platforms (Qidian, Fanqie, Jinjiang, Qimao, Ciweimao). It runs a five-phase pipeline — confirm platform and direction, acquire real ranking data (script scraping first, user-supplied second, built-in knowledge last), analyze genre distribution/word-count/title patterns/new-element signals, output a structured market report (genre heat ranking, new-genre signals, classic-genre dynamics, word-count medians, tag buzzwords, worth-watching directions), and turn findings into topic decisions with hard feasibility rules (sparse data caps feasibility at "medium"). Core belief: one ranking is a clue, only repeated cross-sample patterns are a signal. MIT-licensed. Honest notes: the entire skill is written in Chinese and targets Chinese platforms — irrelevant if you don't write for that market; ranking data goes stale fast, and built-in-knowledge mode must be labeled as hypothesis-only until re-scanned. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Quick answer
story-long-scan is a web-novel market analyst for Chinese platforms Qidian, Fanqie, Jinjiang, Qimao and Ciweimao, written in Chinese. It runs a five-phase pipeline: confirm platform and direction, acquire ranking data by script scraping first, user data second and built-in knowledge last, analyze genre, word-count, title and new-element signals, then output a market report with genre heat rankings and new-genre signals.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 18, 2026📌 Pinned version: commit 7408014Step 0, before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: Node.js (the bundled ranking scrapers and the aggregation script run with node); a Chrome session via this repo's companion browser-cdp skill for the platforms and chart types that need a browser (Qidian's default charts do not); real ranking data, scraped by the scripts, pasted or supplied by you, or, as a last resort, the skill's built-in knowledge; the skill is written in Chinese Install "story-long-scan: Chinese web-novel market rank scanning (Qidian, Fanqie, Jinjiang…)" for me. It gives my agent @zenstory-ai's five-phase market analyst for Chinese long-form web novels: it confirms the platform and direction with you, acquires real ranking data (script scraping first, author-supplied data second, built-in knowledge last), aggregates and analyzes genre distribution, word counts, title patterns and new-element signals across repeated samples, outputs a structured market report (genre heat ranking, new-genre signals, classic-genre dynamics, word-count medians, tag buzzwords, directions worth watching), and turns the findings into 2 to 3 topic decisions whose feasibility is capped by hard rules in its topic-decision reference, all under one core belief: a single ranking is only a clue, and repeated patterns across recent charts and works are what support a trend judgement. MIT-licensed. Repository: https://raw.githubusercontent.com/zenstory-ai/oh-story-claudecode/74080143dd1ac9a39a930fafd99a0dbbe61b946a/skills/story-long-scan/SKILL.md (the skill is the whole skills/story-long-scan/ folder at that commit: SKILL.md plus its references/ subfolder, with references/genre-trends.md, references/topic-decision.md and the five references/platform-*.md files, and its scripts/ subfolder, with scripts/aggregate-rank.js, scripts/cdp-utils.js and the five platform rank scrapers) Pinned version: install the code at commit 74080143dd1ac9a39a930fafd99a0dbbe61b946a exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. Updated October 18, 2026: Skill Harbor regenerated this install prompt from the current upstream file. Upstream restructured the skill around sources and sampling (SKILL.md went from 188 lines down to 160 lines). Phase 1 now reuses the conditions you already gave and asks for the missing ones all at once, with options: platform, channel for Fanqie and Qimao (male, female, or both), rising new books (more new-book samples) versus the mature market (default charts only), and a genre direction or a full overview. Phase 2's data sources were renamed and reworked: script scraping first, author-provided data second, built-in knowledge last; each scan writes into a fresh dated output folder (扫榜/{YYYYMMDD}/); only the chosen platform's reference file is read; and sampling for new character settings, entry settings and tropes goes through the aggregation script's --sample mode, one direction at a time, deduplicated by title and author per platform. An author-supplied link is no longer fetched directly: because Fanqie and Qidian fight scrapers, the skill now asks the author to copy the page content or send a screenshot instead. Phase 3 now treats top back-catalogue hits as existing readership, not as proof a new book can rise, and notes that Fanqie is sampled equally per genre, so its title counts are sampling caps and heat is judged by the median. The hard feasibility rules (sparse data, a first scan with nothing to compare against, or built-in-knowledge mode cap feasibility at "medium") moved out of SKILL.md into references/topic-decision.md, where they still apply. Two reference files were removed, references/reader-profiling.md and references/publishing-guide.md, and the platform quick-reference table and the language section were dropped from SKILL.md, so install the whole folder as it now stands, not SKILL.md alone. Correction: the previous prompt's safety check spoke of allowed hosts declared in SKILL.md; neither the old nor the new SKILL.md declares any allowed hosts list. 1. Fetch the whole skills/story-long-scan/ folder at the pinned commit, SKILL.md, every file under references/ and every file under scripts/, into a temporary folder, then compare SKILL.md with the fingerprint from Skill Harbor's review of 2026-10-18: - SKILL.md: SHA-256 9c30fbc326aa6bed48e83c6c4dff449143f770f0b0e7a77f5e4a953e564aafe1 If it does not match, it changed since the review: do not install it yet, tell me what you found and let me decide. 2. Safety check: read SKILL.md, its reference files and its scripts and review them for anything suspicious (unexpected network calls, shell commands, credential harvesting); STOP on any red flag and tell me. Tell me in plain words what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Confirm these points from the files: - Data comes from three sources, in this order: the bundled scripts scrape the chosen platform's public ranking pages; data the author pastes, screenshots or already has in files; and, only when neither works, built-in knowledge from references/genre-trends.md, which the report must label as hypotheses to verify, with a list of the charts to re-scan. Nothing is scraped or sent anywhere during installation itself. - The scripts connect to the Chinese platforms themselves (Qidian, Fanqie, Jinjiang, Qimao, Ciweimao) when the skill is used; some charts need a Chrome session started through the repo's companion /browser-cdp skill, while Qidian's default charts are read from its mobile pages without Chrome. If a captcha appears, the script stops and waits for the author to complete it manually in Chrome (up to 120 seconds). - Platform heat means something different on each platform (Fanqie counts readers in progress, Qidian its paid charts and total recommendations, Jinjiang favourites, nutrient fluid and points, Qimao heat); completion rates, read-through and subscription conversion are author-backend data the scripts cannot collect, so the skill draws no conclusions from them unless the author provides them. - Every scan writes into a fresh dated folder, 扫榜/{YYYYMMDD}/ (a second scan the same day gets a -2 suffix), never appending to an old scan folder; only the chosen platform's reference file is read; the main session reads the aggregate produced by scripts/aggregate-rank.js first, and sampling runs one direction at a time with --sample, deduplicated by title plus author on the same platform. - The report is shown to the author and written into that folder's 选题决策.md file as its scan-conclusions section; topic decisions follow references/topic-decision.md, and feasibility is capped at "medium" when the data is sparse (fewer than 15 same-direction samples, 10 on Ciweimao), when it is a first scan with no previous period to compare, or in built-in-knowledge mode. 3. When I say yes, install it as a skill: copy SKILL.md, its reference files and its scripts into the agent's skills directory, in a folder named "story-long-scan", keeping the references/ and scripts/ subfolders as they are. 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. tell the agent which platform to scan, Qidian, Fanqie, Jinjiang, Qimao or Ciweimao, plus the channel and the new-book or mature-market focus where they apply; be ready to paste ranking data or complete a captcha if scraping fails; the skill is written in Chinese, so work in Chinese when invoking it). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- 7408014 · 2026-10-18
- Sandbox install test
- Install ran in our test sandbox · Nov 18, 2026 · Observed during the install test: registry.npmjs.org
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.