NCM Listening Profile skill
Skill that pulls your NetEase Cloud Music playlists and listening rankings and prepares prompts for an AI portrait of your taste
- 135 ★ on the repository
- Last commit
- MIT-0
Extended access
Why this level
- Reads data from the logged-in NetEase Cloud Music client through a local debug port and writes it to files on disk
- The finished prompt contains listening history, so passing it to a cloud model sends that data off the machine
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/ncm-listening-profileDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reads data from the logged-in NetEase Cloud Music client through a local debug port and writes it to files on disk
- The finished prompt contains listening history, so passing it to a cloud model sends that data off the machine
This is third-party code. Review the repository files before installing.
What it does
The skill connects to the NetEase Cloud Music desktop client on your computer. It exports a chosen main playlist, the last-week ranking and the all-time ranking into `result/*.jsonl` files. A script next to it precomputes 82 metrics across 10 analysis areas and stores them in `aggregate.json`. The output is two versions of an analysis prompt with local paths filled in. You hand them to the agent or another model, which then describes long-term preferences, recent mood, imagery in titles and rare outliers. Everything lands in a timestamped `outputs/` folder.
Who it is for. NetEase Cloud Music users on Windows or macOS who want to analyse their own listening history with AI.
Good fit when
- A large main playlist has built up and you want to see which songs have stayed in it for years
- You want a last-week snapshot: how many plays and which tracks repeat most
- You want the analysis results to stay on your own machine and be handed to a model piece by piece
- You compare long-term and short-term taste: what stays in the all-time top and what arrived recently
Not a fit when
- You listen in another service: the skill reads only the NetEase Cloud Music client
- You have no desktop client, for example only the mobile app
- You want a profile of someone else: the skill is built for the account owner's own data
Example request
Build my NetEase Cloud Music listening profile from the main playlist and show what stays in the top for yearsLimitations
- It runs on Windows 10 and 11 with client 3.0.0 Beta or newer, and on macOS arm64 with 3.1.7 or newer. The author did not test other systems.
- You need Python 3.10 or newer and an installed NetEase Cloud Music client with an active login.
- Collection runs through the local debug port `127.0.0.1:9222` inside the client page, so a client update can break the script.
- Prompts and documentation are written in Chinese, with an English README in a separate file.
- The quality of the portrait depends on the model you give the prompt to.
How to disable. Delete the skill folder, for example `~/.claude/skills/ncm-listening-profile`, and the `outputs/` directory with exported data. The skill does not modify the NetEase Cloud Music client.
FAQ
What data does the skill collect?
The chosen main playlist, the last-week ranking and the all-time ranking. From them it computes 82 metrics across 10 analysis areas.
Where do the results go?
Into a local `outputs/YYYYMMDD-HHMMSS/` folder. According to the author, the diagnostic log keeps only stages and aggregate statistics.
How do I install the skill in Claude Code?
Run `npx skills add dragonknightk/ncm-listening-profile`. ClawHub, the LobeHub CLI and cloning the repository also work.
What if collection does not start?
Run `python scripts/collect_ncm_profile.py --check` for a self-check that shows whether the client is reachable. The collection log sits in `log/collection_diagnostics.json`.
Is an API key needed?
No, the skill talks to the local client. A key is needed only by the model you give the finished prompt to.
Discussion
No questions or reviews yet.
Related
SkillsA skill that audits accounts and viral posts on Xiaohongshu, Douyin, Kuaishou, WeChat Channels and Bilibili, then drafts a copy.
A skill for daily creator briefings: scans sources, filters fresh signals and drafts posts in your voice.
A skill that teaches agents Chinese official writing from a 1.02M-character corpus: seven genres and parameter checks.
A skill that audits an Amazon listing across CDQ, A9, COSMO and Alexa and returns a JSON report without rewriting the copy
A skill builds an Amazon listing from a brand and competitor ASINs into Excel with title, bullets and search terms.
A skill that builds a panel of 5 to 15 composite customers and tests a landing page, price or email before launch.
README in short
The two-language page opens with the story behind the project and sample paragraphs from a finished portrait. It then lists the 10 analysis areas and 82 metrics, a three-step install and the verified client versions.