macro-data-brief

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to collect, analyze, and report on macroeconomic data (PMI, CPI, social financing, etc.) for investment purposes. It uses a specific data provider (gildata-aidata) and standard web searching to retrieve this public information.
  • [COMMAND_EXECUTION]: While the skill description mentions 'calls' and 'queries' to tools like gildata-aidata and web_search, these are standard agent tool invocations within a sandboxed environment and do not involve arbitrary shell command execution or privilege escalation.
  • [DATA_EXFILTRATION]: No patterns of sensitive data access (e.g., credentials, SSH keys, or environment variables) were detected. The data being processed is public economic data, and the output is provided back to the user in a Markdown report format.
  • [PROMPT_INJECTION]: The skill uses clear instructions for the AI's behavior and does not contain any patterns attempting to bypass safety filters or override system instructions.
  • [REMOTE_CODE_EXECUTION]: There are no indicators of remote code downloads, script execution (like curl | bash), or dynamic code evaluation. All logic is contained within the provided instructional text.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 06:40 AM
Security Audit — agent-trust-hub — macro-data-brief