flipswitch-setup

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill instructs users to install a third-party MCP server from 'https://mcp.flipswitch.io/mcp' and install various SDKs (e.g., '@flipswitch-io/sdk', 'flipswitch-sdk'). These sources are outside the trusted scope, introducing supply-chain risks.\n- [REMOTE_CODE_EXECUTION] (HIGH): The skill exhibits a dangerous indirect injection surface by fetching code snippets from a remote MCP tool ('get_sdk_setup_snippet') and writing them directly to the filesystem (Step 9) without sanitization or review. \n
  • Ingestion point: Tool output from 'mcp__flipswitch__get_sdk_setup_snippet'. \n
  • Capability inventory: 'Write' tool for file creation, 'Bash' for package installation. \n
  • Sanitization: None; the agent writes the remote API response directly to project files. \n- [COMMAND_EXECUTION] (MEDIUM): Utilizes 'Bash' and 'Write' tools to modify the local environment and install dependencies based on external instructions.\n- [DATA_EXFILTRATION] (LOW): Retrieves and processes SDK API keys from a remote server. While functional for the service, it involves handling sensitive credentials via an untrusted third-party endpoint.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 10:58 AM