lazycat-movie-search

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local bash scripts scripts/yts-api.sh and scripts/vuetorrent-api.sh to perform movie searches and interact with a local torrent client.
  • [EXTERNAL_DOWNLOADS]: The yts-api.sh script fetches movie metadata and torrent information from an external API endpoint (https://movies-api.accel.li/api/v2).
  • [CREDENTIALS_UNSAFE]: The vuetorrent-api.sh script contains hardcoded default credentials (admin / adminadmin). These are provided as functional defaults for the target application and do not indicate malicious harvesting.
  • [PROMPT_INJECTION]: The skill processes untrusted movie metadata from an external API, creating an indirect prompt injection surface when displaying results to the user.
  • Ingestion points: Movie titles and details retrieved from the YTS API results in scripts/yts-api.sh.
  • Boundary markers: Not present; external content is displayed directly to the user for selection.
  • Capability inventory: Network access for metadata retrieval and management of local torrent tasks via scripts/vuetorrent-api.sh.
  • Sanitization: Uses jq for basic structured data parsing, but does not apply specific sanitization for malicious instruction patterns within the metadata text.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 10:04 AM
Security Audit — agent-trust-hub — lazycat-movie-search