keyword-research

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or security vulnerabilities were identified in the analyzed files.
  • [DATA_EXFILTRATION]: The skill is designed to record research deliverables and strategy decisions to local paths within the memory directory (e.g., memory/research/). This data movement is transparent, localized, and fundamental to the skill's utility in persistent SEO strategy.
  • [PROMPT_INJECTION]: The skill processes user-provided inputs such as seed keywords and competitor URLs. While this represents a surface for indirect prompt injection, the workflow is analytical and lacks high-privilege capabilities that could be abused via injected instructions. (1) Ingestion points: User topics, seed keywords, and market inputs (SKILL.md). (2) Boundary markers: Absent. (3) Capability inventory: Local file system writes to the memory directory; no evidence of shell execution, dynamic code evaluation, or direct network operations in the provided instructions. (4) Sanitization: No specific input sanitization or validation routines defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 07:37 PM