seo-cluster

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the claude-seo command-line utility to perform specialized tasks.
  • It runs render_page.py for URL fetching and processing.
  • It executes dataforseo_costs.py to manage API usage costs.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface (Category 8).
  • Ingestion points: External data enters the context through user-provided seed keywords (SKILL.md) and organic search results (references/serp-overlap-methodology.md) harvested from Google.
  • Boundary markers: Collected data is structured into a cluster-plan.json file and passed as a structured JSON object (cluster_context) when invoking downstream skills.
  • Capability inventory: The skill has the ability to write multiple files to the workspace, modify existing markdown files to inject internal links (references/execution-workflow.md), and trigger external content generation skills.
  • Sanitization: The skill relies on render_page.py which includes SSRF protection mechanisms (url_safety).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 09:13 AM
Security Audit — agent-trust-hub — seo-cluster