Multi-Platform Publisher
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources and processes it through multiple capability tiers, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill reads source content from local files (e.g.,
article.md) and metadata (titles, tags) provided by the user or extracted from external documents. - Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands within the ingested content.
- Capability inventory: The skill executes shell commands (
wechatsync,biliup), makes network requests viacurlto a local daemon (xhs-mcp), and executes Python code (bilibili-api-python). - Sanitization: No sanitization or validation of the article content is performed before it is passed to adaptation agents or command-line tools.
- [COMMAND_EXECUTION]: The skill constructs shell commands by interpolating variables such as article titles and content. If the underlying platform does not properly escape these values, a malicious article title containing shell metacharacters could lead to arbitrary command execution on the host system.
- [CREDENTIALS_UNSAFE]: The skill provides templates and instructions for users to manually extract and provide sensitive session credentials (
sessdata,bili_jct,buvid3) from browser cookies to the Bilibili API. While no secrets are hardcoded in the skill itself, this workflow encourages users to handle raw session tokens directly in scripts. - [EXTERNAL_DOWNLOADS]: The skill relies on and recommends the installation of several third-party tools and libraries from GitHub repositories (e.g.,
wechatsync/Wechatsync,xpzouying/xiaohongshu-mcp,biliup/biliup). These resources are hosted outside of verified organizations and represent a supply chain dependency risk. - [DYNAMIC_EXECUTION]: The skill uses Python script generation to interact with the Bilibili API and utilizes
curlto interact with a local API daemon. It also uses string interpolation to build executable shell commands from user-supplied or file-based content.
Audit Metadata