ai-seo-report

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources via MCP tools (aiseo_list_projects, aiseo_get_scan, aiseo_list_prompts, aiseo_list_sources, aiseo_list_competitors, aiseo_list_snippets) and incorporates this data into its report output.
  • Ingestion points: Data returned by the MCP endpoint at aiseotracker.com/api/agent/mcp is used to build the REPORT DATA block in SKILL.md.
  • Boundary markers: No specific delimiters or "ignore instructions" directives are provided to prevent the agent from obeying potential instructions embedded in project names, search snippets, or competitor data.
  • Capability inventory: The skill uses MCP tools for data retrieval and instructs the agent to fetch remote content from the author's domain.
  • Sanitization: There is no evidence of input validation or sanitization for the content retrieved from the tools before it is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The skill contains instructions to fetch its own configuration or setup files from external URLs.
  • Evidence: The description and body instruct the agent to fetch aiseotracker.com/ai-seo-report/SKILL.md and aiseotracker.com/SKILL.md for initialization and integrity checks.
  • Note: These downloads target the author's own domain and are consistent with the skill's management procedures.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 04:49 PM