bilibili-viral-topic

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on a local Python script scripts/search_bilibili_viral_topic.py to handle data collection, scoring, and output formatting.
  • [EXTERNAL_DOWNLOADS]: Performs network requests to api.bilibili.com to retrieve video metadata, search results, and ranking lists.
  • [DATA_EXPOSURE]: The script reads the BILIBILI_COOKIE environment variable if provided by the user to mitigate rate-limiting on Bilibili's platform.
  • [DATA_EXPOSURE]: Manages a local JSON cache at ~/.cache/account-growth/bilibili_followers.json to store follower counts for up to 24 hours.
  • [INDIRECT_PROMPT_INJECTION]: Processes untrusted data (video titles, author names, and descriptions) from the Bilibili API which is then presented to the user or agent.
  • Ingestion points: Data is fetched via fetch_search and fetch_rankings in scripts/search_bilibili_viral_topic.py.
  • Boundary markers: No explicit delimiters are used to separate API content from the prompt context.
  • Capability inventory: The script has permissions for network operations (GET requests) and local file system writes (cache file).
  • Sanitization: Uses a strip_html function to remove basic HTML tags from API responses.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 02:54 PM
Security Audit — agent-trust-hub — bilibili-viral-topic