ds-seo-weekly

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses the dynamic context syntax to load marketing context from the local path .agents/product-marketing-context.md. This occurs when the skill is loaded and is used for personalization.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (Google Search Console and GA4) which can be provided manually or via tools.
  • Ingestion points: User-provided SEO data in CSV, TSV, or JSON format (Step 2, Path B).
  • Boundary markers: Absent; data is passed directly to the processing script.
  • Capability inventory: Execution of a local Python script (ds_utils.py) via the Bash tool.
  • Sanitization: The data is validated and processed by ds_utils.py before being interpreted by the agent, reducing but not eliminating the injection surface.
  • [COMMAND_EXECUTION]: The skill executes a local Python script (ds_utils.py) using the Bash tool to perform specialized SEO calculations. This tool usage is restricted by the skill configuration to only allow the specific utility script.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 03:05 AM