report-builder
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is comprised of natural language instructions and YAML metadata with no executable components, scripts, or network exfiltration risks.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user data to build reports, it does not provide any tools or scripts that could be exploited through injection. 1. Ingestion points: Processes data for business reports, research summaries, and dashboards as described in SKILL.md. 2. Boundary markers: No specific delimiters are defined in the instructions for untrusted data. 3. Capability inventory: No tools, scripts, or system-level capabilities are defined. 4. Sanitization: Guidelines explicitly require that the agent validate all inputs before processing.
Audit Metadata