local-seo
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (
campaign-tracker.py) using interpolated brand identifiers (--brand {slug}). This creates a potential command injection surface if the input variable is not strictly validated.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests configuration data from local brand profiles and guidelines to customize its output and behavior.\n - Ingestion points: Reads
profile.jsonand_manifest.jsonfrom the~/.claude-marketing/brands/directory (referenced inSKILL.md).\n - Boundary markers: None identified in the prompt templates or instructions.\n
- Capability inventory: Includes subprocess execution for campaign management tasks.\n
- Sanitization: No explicit validation or sanitization of the ingested file content is documented.\n- [SAFE]: References official industry resources from a well-known service (Google's official blog) for 2026 booking trends and updates.
Audit Metadata