semrush-tool
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a purely instructional guide for SEO workflows using Semrush and does not contain any executable code or sensitive file access.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze potentially untrusted data from external Semrush reports. 1. Ingestion points: Data enters the agent context via the Semrush MCP server outputs, including competitor organic research and site audits (SKILL.md). 2. Boundary markers: The instructions lack explicit markers to differentiate between the agent's instructions and the external SEO data being analyzed. 3. Capability inventory: No direct file writing or subprocess execution is initiated by this specific skill file. 4. Sanitization: There are no instructions for sanitizing or validating the input retrieved from Semrush before it is processed by the agent.
- [METADATA_POISONING]: The metadata fields are consistent with the content and do not attempt to deceive the analyzer or the agent regarding the skill's capabilities.
Audit Metadata