seo-google

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFENO_CODECOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The analyzed files consist only of markdown documentation and templates; the functional Python scripts referenced in the commands are not present in the provided source set.
  • [SAFE]: Analysis of the instructions and references shows legitimate use of official Google APIs for SEO data collection with no signs of malicious intent or data exfiltration.
  • [COMMAND_EXECUTION]: The skill is designed to trigger local Python scripts that perform search analytics, indexing requests, and performance audits.
  • [CREDENTIALS_UNSAFE]: Security best practices are followed for managing API keys and service account JSON files by guiding users to store them in a secure local directory (~/.config/claude-seo/).
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from YouTube comments and webpage text. Ingestion points: nlp_analyze.py, youtube_search.py. Boundary markers: Markdown report templates. Capability inventory: File-write (reports), Network-ops (Google APIs). Sanitization: No sanitization logic was provided for review.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 08:13 PM
Security Audit — agent-trust-hub — seo-google