skills/teamday-ai/agents/seo/Gen Agent Trust Hub

seo

Fail

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: HIGHCREDENTIALS_UNSAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The script scripts/umami-report.ts includes a hardcoded token value for UMAMI_SHARE_ID ("GRVFnQSQSLlfRjCa"). This token is used to authenticate requests to the Umami Analytics API as a fallback if no environment variable is provided.
  • [DATA_EXFILTRATION]: The skill performs outbound network requests to https://umami.lsd.sk to retrieve site analytics data.
  • [DATA_EXFILTRATION]: The utility scripts/health-score.ts crawls the website www.teamday.ai (the vendor's domain) by fetching sitemaps and individual page content to analyze SEO health.
  • [COMMAND_EXECUTION]: The audit script scripts/health-score.ts executes the shell command find via Bun's execSync to locate and list HTML files on the local filesystem.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface in scripts/health-score.ts which ingests and parses untrusted HTML content from the web.
  • Ingestion points: External HTML content fetched from URLs in the sitemap.
  • Boundary markers: Absent; the parsed data is presented to the agent without delimiters.
  • Capability inventory: The skill has access to Bash and Write tools.
  • Sanitization: Absent; the script extracts text from tags via regular expressions without sanitization.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 21, 2026, 09:52 PM
Security Audit — agent-trust-hub — seo