event-follow-up-engine
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill proactively addresses the risk of indirect prompt injection by instructing the agent to treat all fetched pages, uploaded files, and tool outputs strictly as data and never as instructions. It includes a specific protocol to report and ignore any embedded commands found in external content.
- [COMMAND_EXECUTION]: The skill utilizes a local script,
scripts/text_check.py, to perform text validation tasks such as character counting and date verification. This script uses only Python's standard library and is executed in a controlled manner by the agent to ensure output quality. - [DATA_EXPOSURE_AND_EXFILTRATION]: The instructions explicitly forbid writing to any CRM or marketing platforms. While it may read data from connected platforms at the user's request, it enforces a read-only policy to prevent unauthorized data modification or exfiltration.
- [SAFE]: The skill follows established security best practices, including a 'comprehension gate' that requires user confirmation before proceeding with complex tasks and 'Anti-Hallucination Rules' that prevent the fabrication of data or sources.
Audit Metadata