bridge-conflicts

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes information from external sources such as MCP client configuration files and system port diagnostics. This ingestion creates a surface for indirect prompt injection if an attacker can control the content of these files or the output of diagnostic tools.\n
  • Ingestion points: The skill reads multiple external configuration files including mcp-servers.json (Ollama, LM Studio), settings.json (Zed, Claude Desktop), .cursor/mcp.json (Cursor), and various other application settings. It also processes output from port analysis commands like lsof and netstat.\n
  • Boundary markers: There are no explicit boundary markers or instructions defined to prevent the agent from following potential instructions embedded within the configuration data.\n
  • Capability inventory: The skill identifies conflicts that require file access and diagnostic execution. The agent context typically includes capabilities for file system interaction and shell command execution.\n
  • Sanitization: No sanitization or validation mechanisms are specified for the data retrieved from the configuration files or diagnostic tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 03:10 PM