SEMrush Automation
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a standard implementation for SEO automation and does not exhibit any malicious behavior, credential harvesting, or persistence mechanisms.
- [EXTERNAL_DOWNLOADS]: The skill references external configuration endpoints, including the MCP server at
https://rube.app/mcpand documentation athttps://composio.dev/toolkits/semrush. These URLs are associated with the vendor's infrastructure for the provided integration. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the SEMrush API, which serves as an untrusted data source.
- Ingestion points: SEMrush tool outputs in SKILL.md (e.g.,
data/keyword_datafromSEMRUSH_DOMAIN_ORGANIC_SEARCH_KEYWORDS). - Boundary markers: None defined.
- Capability inventory: Tool access is limited to read-only SEO analysis metrics.
- Sanitization: No explicit sanitization or filtering of external API content is mentioned.
Audit Metadata