oc-thumbnail-artist

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface via untrusted input.\n
  • Ingestion points: The skill ingests untrusted title and caption data from external sources or prior agent steps (SKILL.md).\n
  • Boundary markers: No boundary markers or instructions to ignore embedded instructions are present to protect the shell command context.\n
  • Capability inventory: The skill utilizes the Bash tool to execute shell commands (SKILL.md).\n
  • Sanitization: No sanitization, escaping, or validation logic is defined for the input variables before they are interpolated into shell commands.\n- [COMMAND_EXECUTION]: Potential command injection vulnerability.\n
  • The skill instructs the agent to construct bash commands by interpolating variables directly into a string: oc --project <P> thumbnail --input <VIDEO> --start <S> --end <E> --aspect 16:9 --title "<hook headline>".\n
  • An attacker or a compromised upstream agent providing a <hook headline> containing shell metacharacters (e.g., $(id), ; rm -rf /, or backticks) could achieve arbitrary command execution within the environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 03:48 AM
Security Audit — agent-trust-hub — oc-thumbnail-artist