douyin-share-info
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute a local TypeScript script (
scripts/main.ts) using thebunruntime vianpx. - [CREDENTIALS_UNSAFE]: The script
scripts/main.tscontains logic to read environment variables from potentially sensitive file paths, including~/.baoyu-skills/.envand the current working directory. - [DATA_EXFILTRATION]: The script transmits user-provided URLs and the
TIKHUB_API_KEYto an external API service (defaulting toapi.tikhub.io). While this is functional, it involves sending credentials to a third-party domain. - [PROMPT_INJECTION]: The skill processes untrusted external content (video metadata and descriptions) from the TikHub API, which constitutes an indirect prompt injection surface.
- Ingestion points: External data is fetched via the
fetchByShareUrlfunction inscripts/main.ts. - Boundary markers: None are present in the
SKILL.mdor prompt templates to delineate external data for the agent. - Capability inventory: The script can perform network requests and write data to the local file system (via the
--rawCLI option). - Sanitization: The script parses the API response as JSON and extracts specific fields but does not perform content-level sanitization of the strings returned.
Audit Metadata