baoyu-danger-x-to-markdown
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: Includes a hardcoded public Bearer token in
scripts/constants.ts. This token is a known credential used for guest access to the X API. - [COMMAND_EXECUTION]: Uses
bunornpxto execute the skill's TypeScript scripts. It also launches browser processes (such as Google Chrome or Microsoft Edge) to programmatically extract authentication cookies and uses system utilities likewslpathorcmd.exefor environment-specific path resolution. - [EXTERNAL_DOWNLOADS]: Fetches data and media assets from X (x.com, twitter.com, twimg.com). These operations are directed at a well-known service and are central to the skill's primary function.
- [PROMPT_INJECTION]: The skill processes untrusted text from external tweets and articles, which represents a surface for indirect prompt injection. * Ingestion points: Reads content from X URLs (processed in
scripts/graphql.ts). * Boundary markers: Uses YAML front matter and Markdown headers, but lacks specific instructions to the agent to disregard commands found within the external content. * Capability inventory: Includes file system writes, network requests, and local script execution. * Sanitization: Content is formatted into Markdown without filtering for potential prompt injection patterns.
Audit Metadata