Skill Trust Decision
guaikei-douyin-trending-videos-and-rankings
This is a legitimate Douyin data analysis skill but contains two significant documentation violations: it prints promotional contact info when the token is invalid (violating its own 'neutral error only' rule), and it writes logs to disk despite claiming no filesystem writes. Both issues are doc-to-code mismatches rather than malicious behavior.
Most direct threat evidence
High Doc Mismatch
Promotional content printed on token error When GUAIKEI_API_TOKEN is invalid, the code prints '快速获取通道: www.guaikei.com' which is promotional contact info. SKILL.md line 52 explicitly prohibits outputting any contact info or promotion on token errors.
src/utils/token.js:22 Why this conclusion was reached
2/4 dimensions flagged Block
Declared vs actual capability
1 undeclared or violating capabilities were inferred.
Review
Hidden execution and egress
13 lower-risk artifacts were extracted and still need context.
Block
Attack chain and severe findings
The report includes 0 attack-chain steps and 1 severe findings.
Pass
Dependencies and supply chain hygiene
Dependencies are present but no obvious high-risk issue stands out.
What drove the risk score up
Doc deception - promotional output +20
token.js line 23 prints '快速获取通道: www.guaikei.com' when token is invalid, directly violating SKILL.md line 52 which prohibits outputting contact info/links on token error
Undeclared filesystem write +15
SKILL.md claims '不写入文件系统', but log.js writes output files to ./logs/ directory - this is a hidden filesystem:WRITE operation not declared in docs
No shell execution +-5
Clean - only uses Node.js built-in modules, no subprocess/exec
No credential theft +-5
Token handled properly via env var, no harvesting of other credentials
Legitimate network behavior +-5
All network requests go to www.guaikei.com API, no suspicious IPs or C2 communication
Most important evidence
High Doc Mismatch
Promotional content printed on token error
When GUAIKEI_API_TOKEN is invalid, the code prints '快速获取通道: www.guaikei.com' which is promotional contact info. SKILL.md line 52 explicitly prohibits outputting any contact info or promotion on token errors.
src/utils/token.js:22 Remove the printSuccess line or replace with neutral message. The doc says: '令牌无效时仅输出中性错误信息，不输出任何联系方式、官网链接或推广内容'
Medium Doc Mismatch
Undeclared filesystem write operation
SKILL.md claims '不写入文件系统（日志除外）' but the logging is not declared as a filesystem:WRITE permission. The skill writes JSON results to ./logs/ directory via log.taskWrite().
src/utils/log.js:30 Update SKILL.md to declare filesystem:WRITE permission for log output, or clarify that logging is a declared exception
Low Doc Mismatch
Error message contains contact info
src/utils/request.js line 76 includes WeChat contact '13395823479' in AuthError message when 401/403 status received from API.
src/utils/request.js:76 This is a fallback error from the external API, not printed by the skill itself on token validation failure. However, it could still expose users to contact harvesting. Consider wrapping or filtering API error messages.
Declared capability vs actual capability
Network Pass
Declared READ
→ Inferred READ
src/config/constants.js:10 - BASE_URL: www.guaikei.com Filesystem Block
Declared NONE
→ Inferred WRITE
src/utils/log.js:30 - fs.promises.writeFile writes to ./logs/ Shell Pass
Declared NONE
→ Inferred NONE
No subprocess/exec usage found Environment Pass
Declared READ
→ Inferred READ
process.env.GUAIKEI_API_TOKEN used in all CLI files Suspicious artifacts and egress
Medium External URL
https://www.douyin.com/user/MS4wLjABxxx SKILL.md:108
Medium External URL
https://www.douyin.com/video/xxx SKILL.md:111
Medium External URL
http://json-schema.org/draft-07/schema# assets/comment_cli_req.schema.json:2
Medium External URL
https://www.guaikei.com readme.md:24
Medium External URL
https://v.douyin.com/xxx references/options.md:48
Medium External URL
https://clawhub.ai/user/engheng-art skill-card.md:9
Medium External URL
https://clawhub.ai/engheng-art/skills/guaikei-douyin-trending-videos-and-rankings skill-card.md:43
Medium External URL
https://www.douyin.com/user/$ src/api/comment.js:16
Medium External URL
https://www.douyin.com/note/xxx src/douyin/comment-cli.js:37
Medium External URL
https://www.douyin.com/note/ src/validate/comment.js:10
Medium External URL
https://www.douyin.com/video/ src/validate/comment.js:12
Medium External URL
https://www.douyin.com/user/ src/validate/post.js:10
Dependencies and supply chain
| Package | Version | Source | Known vuln | Notes |
|---|---|---|---|---|
| Node.js built-in only | >=16.14.0 | runtime | No | No external npm dependencies - uses only built-in modules (fs, path, https, querystring) |
File composition
35 files · 2765 lines
JavaScript 21 files · 1803 linesMarkdown 5 files · 536 linesJSON 9 files · 426 lines
Files of concern · 6
src/utils/token.js Promotional content printed on token error
readme.md https://www.guaikei.com
SKILL.md https://www.douyin.com/user/MS4wLjABxxx · https://www.douyin.com/video/xxx
src/utils/request.js Error message contains contact info
src/douyin/comment-cli.js https://www.douyin.com/note/xxx
references/options.md https://v.douyin.com/xxx
Other files · search-cli.js · post-cli.js · keyword.js · args.js · search_cli_resp.schema.json · search.js
Security positives
No shell command execution - only Node.js built-in modules used
No credential harvesting beyond the expected API token
Input validation with keyword sanitization prevents injection attacks
No base64 encoded execution or obfuscation
No sensitive path access (~/.ssh, ~/.aws, .env)
No remote script execution (curl|bash, wget|sh)
No direct IP network requests or C2 communication
Clean dependency tree - no external npm packages
Token validation uses regex for format checking (16-256 chars, alphanumeric+underscore+dash)
Error handling is comprehensive with proper exit codes