competitive-analysis

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and referenced files are entirely focused on providing a structured methodology for business intelligence and do not contain any malicious patterns or attack vectors.
  • [PROMPT_INJECTION]: No evidence of prompt injection or instructions to bypass safety guidelines were found. The 'Handling Incomplete Prompts' section correctly instructs the agent to ask clarifying questions rather than generating fake data.
  • [DATA_EXFILTRATION]: There are no attempts to access sensitive local files or transmit data to unauthorized external endpoints.
  • [REMOTE_CODE_EXECUTION]: The skill does not include any commands for downloading or executing external scripts, and no third-party package dependencies are declared.
  • [COMMAND_EXECUTION]: No shell commands, privilege escalation attempts, or persistence mechanisms are present in the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 06:07 AM
Security Audit — agent-trust-hub — competitive-analysis