houdini-vex-programming
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for interacting with external Houdini services (Houdini MCP) and tools (
build_network,verify_network). This creates a surface where untrusted user-provided VEX code could influence the agent's behavior during network construction or verification. - Ingestion points: User-provided VEX code snippets, wrangle descriptions, and results read back from the Houdini environment.
- Boundary markers: The skill does not define specific delimiters or instructions to ignore potential injections within the VEX code it processes.
- Capability inventory: The skill utilizes tools for building networks (
build_network), verifying them (verify_network), and searching documentation via an external service (fxhoudinimcpon port 8100). - Sanitization: No explicit sanitization or validation of the VEX code is provided before it is passed to the network building tools.
Audit Metadata