clipcat

Fail

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides installation instructions that involve downloading and executing scripts directly from the vendor's website (curl -fsSL https://clipcat.ai/cli | bash and irm https://clipcat.ai/cli.ps1 | iex). This is a documented installation method for the vendor-provided tool.- [EXTERNAL_DOWNLOADS]: The skill fetches the clipcat CLI and related resources from the author's domain at https://clipcat.ai.- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes external content from TikTok, such as creator posts, product reviews, and video captions, which could contain malicious instructions.
  • Ingestion points: TikTok e-commerce data retrieval commands (e.g., creator posts, product reviews) described in SKILL.md.
  • Boundary markers: Absent. The skill does not define specific delimiters or warnings for the agent when processing retrieved TikTok data.
  • Capability inventory: Remote script execution (installation), network operations (via CLI), and local file system access (configuration storage).
  • Sanitization: Absent. The instructions do not specify any validation or sanitization for data retrieved from external TikTok sources.
Recommendations
  • HIGH: Downloads and executes remote code from: https://clipcat.ai/cli - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 19, 2026, 06:35 AM
Security Audit — agent-trust-hub — clipcat