Analysis of comments on Bilibili videos
v1.0.0B站视频评论与弹幕深度分析。输入B站视频链接或BV号，自动采集评论、弹幕数据，进行情感分析、关键词提取、热度分析，生成结构化分析报告。
⭐ 0· 27·0 current·0 all-time
byUnclecheng@unclecheng-li
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description match the included code: analyze_video.py implements fetching video info, danmaku (XML), and comments from Bilibili and runs Chinese NLP (jieba, SnowNLP). No unrelated credentials, binaries, or exotic network endpoints are requested.
Instruction Scope
SKILL.md limits actions to public Bilibili APIs and local analysis. The code follows that. Minor mismatches: SKILL.md warns unauthenticated comment fetch is limited (~20 hot comments) but analyze_video.py will page up to max_pages (default 50) using the public reply endpoint (and passes a BV string as oid) — this may not work as-documented. The scripts do not read local secrets or unexpected files, and they only call Bilibili endpoints.
Install Mechanism
There is no platform install spec, but the bundle includes helper install scripts that run pip to install dependencies (install_deps.sh and install_check.py). Packages are standard public PyPI packages (jieba, snownlp, bilibili-api-python). No downloads from unknown URLs or archives; however there are inconsistent dependency manifests (requirements.txt vs scripts/requirements.txt) and mismatched import/package names (install_check.py checks for 'bilibili_api' while the shell script installs 'bilibili-api-python').
Credentials
The skill declares no environment variables, no credentials, and the code does not attempt to access env variables or system config paths. All network calls go to Bilibili APIs as expected for the stated purpose.
Persistence & Privilege
Skill is not always-enabled and does not request persistent or elevated privileges. It does not modify other skills or system-wide agent settings.
Assessment
This skill appears to do what it says: fetch public Bilibili video info, danmaku, and comments and run local analysis. Before you run anything: 1) Inspect install_deps.sh and install_check.py — they will call pip to install packages (run them in a virtualenv or container to avoid contaminating your system Python). 2) Note the inconsistent dependency lists (requirements.txt vs scripts/requirements.txt) and the package/import name mismatch ('bilibili_api' vs 'bilibili-api-python'); you may need to reconcile these before installing. 3) The script uses public endpoints and claims no login is required, but unauthenticated comment access may be limited — if you need full comment history the tool may require authenticated API access (not included). 4) If you plan to integrate this as a skill that can be invoked autonomously, be aware the included install scripts will install packages if executed; review them and run manually if you prefer explicit control.Like a lobster shell, security has layers — review code before you run it.
latestvk97c9vs5rx6sv6v249mkhg83tx846qxq
License
MIT-0
Free to use, modify, and redistribute. No attribution required.