geo-tracking

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process information from a local Obsidian vault or knowledge base, creating a vulnerability surface where untrusted data could influence agent behavior.
  • Ingestion points: SKILL.md directs the agent to read notes from a local Obsidian vault (.obsidian directory) to gather brand and product facts for context.
  • Boundary markers: The instructions lack delimiters or safety warnings to ignore potential instructions embedded within the ingested notes.
  • Capability inventory: The skill utilizes file read/write permissions to maintain an SEO action log and references a local audit script (scripts/seo_audit.py).
  • Sanitization: There is no evidence of filtering or sanitizing the content retrieved from the vault before the agent processes it.
  • [EXTERNAL_DOWNLOADS]: The skill fetches configuration files from trusted organizations and well-known services to verify the authenticity of network traffic.
  • Evidence: Downloads IP range JSON files from OpenAI (openai.com), Anthropic (claude.com), and Perplexity (perplexity.com) for crawler verification.
  • [COMMAND_EXECUTION]: The skill refers to the use of an internal script to automate SEO auditing tasks.
  • Evidence: References scripts/seo_audit.py to score pages flagged in visibility reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 06:29 AM