seo-keyword-classifier

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The SKILL.md file includes a setup command using npx skills add pointing to the author's GitHub repository (github.com/peter-tu-zynkr/zynkr-skill-builder). This is a legitimate installation mechanism for the vendor's ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes keyword data from external packets, which is a potential surface for indirect prompt injection. 1. Ingestion points: Data enters Step 1 via SEO_PACKET ▸ KeywordMap. 2. Boundary markers: The instructions do not specify any delimiters or safety warnings for the ingested content. 3. Capability inventory: The skill uses the google-workspace tool for reading rubrics and templates but does not possess high-privilege capabilities like arbitrary command execution or network exfiltration. 4. Sanitization: No validation or sanitization of the keyword strings is mentioned.
  • [SAFE]: The skill's core logic is restricted to SEO text classification and incorporates a mandatory human-in-the-loop checkpoint in Step 4, where the user must approve the grouping version before the skill proceeds to handoff.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:30 PM
Security Audit — agent-trust-hub — seo-keyword-classifier