bilibili-portfolio-search

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits the targeted Bilibili UID and a pagination cursor to the vendor-owned API at https://redfox.hk/story/api/bili/userWorkList. This is a standard functional requirement for the service provided by the skill author.
  • [PROMPT_INJECTION]: The skill processes untrusted external content (Bilibili video titles, tags, and category names). While this presents a potential surface for indirect prompt injection, the skill lacks dangerous capabilities such as file system writing or command execution that could be exploited through this vector.
  • Ingestion points: Video metadata is retrieved via the fetch_user_works function in scripts/fetch_bili_user_works.py.
  • Boundary markers: No specific delimiters or warnings are used when interpolating external content into the AI's context.
  • Capability inventory: The skill possesses no capabilities for subprocess execution (other than the internal script), file-write operations, or network exfiltration of local data.
  • Sanitization: External content is rendered as-is in Markdown tables without explicit sanitization for instruction directives.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 09:46 AM
Security Audit — agent-trust-hub — bilibili-portfolio-search