social-content-planner

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill integrates with a hosted Model Context Protocol (MCP) server at https://apikit.simplified.com/mcp. This allows the agent to utilize social media automation and analytics tools provided by the Simplified platform.\n- [SAFE]: The instructions include mandatory human-in-the-loop verification steps. The agent is required to present all post drafts, selected accounts, and platform settings for explicit user confirmation before executing any schedule or queue operations.\n- [PROMPT_INJECTION]: The skill processes external data (account details and performance analytics) which constitutes an indirect prompt injection surface.\n
  • Ingestion points: Account discovery via social_getSocialMediaAccounts and retrieval of analytics data as described in the workflow.\n
  • Boundary markers: The workflow mandates showing the final posts and obtaining explicit confirmation before execution, creating a clear boundary between data processing and action.\n
  • Capability inventory: The skill can read social account configurations and create, draft, or schedule social media posts.\n
  • Sanitization: Security is maintained through mandatory user review and approval of all generated content and metadata before platform commitment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 09:16 PM
Security Audit — agent-trust-hub — social-content-planner