skills/q-qp-p/pstack/ketchup-video/Gen Agent Trust Hub

ketchup-video

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONPRIVILEGE_ESCALATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (screenshots, recordings, diffs, and data produced by work) to generate slide labels and subtitles. This creates a surface where malicious content within project files could influence the agent's output. \n
  • Ingestion points: SKILL.md (referenced data from screenshots, recordings, and diffs). \n
  • Boundary markers: None specified to protect against embedded instructions. \n
  • Capability inventory: File writing and video rendering tools. \n
  • Sanitization: None mentioned. \n- [COMMAND_EXECUTION]: The instructions require the agent to render media files (e.g., MP4), which involves the use of external tools or scripts. \n- [PRIVILEGE_ESCALATION]: The skill explicitly instructs the agent to treat video rendering and file writing as an exception to standard read-only constraints, effectively requesting a bypass of platform-level security restrictions.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 05:09 PM