gemini-tool-orchestrator

Warn

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to translate user intent into shell commands for over 20 security-focused CLI tools including nmap, nuclei, masscan, and ffuf.
  • [DYNAMIC_EXECUTION]: The skill uses an orchestration pattern where the agent dynamically generates exact commands, flags, and parameters at runtime.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The agent processes raw artifacts and normalized findings from tools that scan external, potentially attacker-controlled targets (e.g., nuclei, katana, httpx).
  • Boundary markers: The instructions do not specify the use of delimiters or ignore-instructions when processing tool output to prevent the agent from following embedded malicious instructions.
  • Capability inventory: The agent has extensive shell execution capabilities across many tools and is instructed to write findings and artifacts to the local filesystem.
  • Sanitization: There are no explicit requirements for the agent to sanitize, validate, or filter data retrieved from external tools before processing it.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 26, 2026, 09:20 PM
Security Audit — agent-trust-hub — gemini-tool-orchestrator