youtube-research

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a bash script (scripts/crawlora.sh) to perform network operations. This script acts as a minimal wrapper for curl, specifically configured to interact with the vendor's API at api.crawlora.net. It includes safe handling of query parameters using curl's encoding features.
  • [DATA_EXFILTRATION]: The skill is designed to transmit a user-provided API key to api.crawlora.net. This communication is limited to the vendor's own infrastructure (crawlora-org) and is required for the skill to function as a YouTube research tool.
  • [PROMPT_INJECTION]: The skill processes untrusted external data, such as YouTube transcripts and user comments, which presents a surface for indirect prompt injection. However, the risk is mitigated as the skill's capabilities are scoped to the vendor's specific API endpoints and it does not possess high-privilege capabilities like arbitrary file system access or uncontrolled network operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 07:16 AM
Security Audit — agent-trust-hub — youtube-research