smart-screenshot-capture

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a legitimate workflow for visual documentation using Model Context Protocol (MCP) tools. No malicious instructions, prompt injection attempts, or obfuscated content were found during the analysis.
  • [COMMAND_EXECUTION]: The instructions explicitly direct the agent to use specific, non-mutating MCP tools (e.g., mcp_winforms_take_screenshot, mcp_winapp_take_screenshot) rather than arbitrary shell commands like screencapture, reducing the risk of command injection.
  • [DATA_EXFILTRATION]: The skill manages screenshot storage within deterministic local directories (e.g., Output/ManualEvidence/). No instructions were found that attempt to transmit captured UI data to external network endpoints or unauthorized domains.
  • [EXTERNAL_DOWNLOADS]: The skill includes an upstream reference URL in its metadata for documentation purposes. It does not perform any runtime downloads, remote script executions, or installations from untrusted sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 10:51 PM
Security Audit — agent-trust-hub — smart-screenshot-capture