baoyu-danger-x-to-markdown
Warn
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/paths.tsexecutescmd.exeandwslpathviaexecSyncto resolve Windows user profile paths when running in WSL.\n- [COMMAND_EXECUTION]: The skill launches a browser (Chrome/Edge) usingbaoyu-chrome-cdpinscripts/cookies.tsto automate authentication and cookie harvesting from X.com.\n- [EXTERNAL_DOWNLOADS]: Thescripts/media-localizer.tsutility downloads media assets from well-known Twitter servers (pbs.twimg.com,video.twimg.com) to save them locally.\n- [DATA_EXFILTRATION]: The skill retrieves authentication tokens from environment variables or local files and sends them to X.com GraphQL endpoints to retrieve content.\n- [PROMPT_INJECTION]: The skill processes untrusted content from X, creating an indirect prompt injection surface. \n - Ingestion points: Tweet and article text fetched from X.com in
scripts/graphql.ts. \n - Boundary markers: None present in the generated markdown output. \n
- Capability inventory: File writing (
writeFile), command execution (execSync), and browser automation via Chrome DevTools Protocol. \n - Sanitization: Basic markdown escaping is applied to content in
scripts/markdown.ts.
Audit Metadata