fxhoudinimcp
Related Servers
Alternatives to fxhoudinimcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to directly control SideFX Houdini, including creating nodes, setting parameters, executing Python, capturing viewports, and rendering frames, via 57 MCP tools.1MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server for SideFX Houdini that enables AI assistants to interactively control, construct, inspect, and visually verify 3D procedural scenes through tools like node creation, parameter editing, viewport capture, and rendering.MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to control SideFX Houdini 3D software, providing tools for scene management, node operations, rendering, and more through the Model Context Protocol.1MIT
- AlicenseAqualityBmaintenanceDrive SideFX Houdini from any agent: build node networks in one all-or-nothing call, cook, render, capture the viewport, and read the offline docs of your Houdini build. 20 tools with modes.20197 PyPI16MIT
- AlicenseNot gradedqualityDmaintenanceEnables natural language control of SideFX Houdini for tasks including node management, parameter editing, and geometry inspection. It leverages RPYC to execute Python scripts and manage scene data through an MCP-compatible interface.MIT
- AlicenseNot gradedqualityCmaintenanceEnables language models to inspect the current Houdini scene, create and modify nodes, and run VEX/Python code in the running Houdini session via MCP.1MIT
TDQS
Scored across 216 tools
Several tool families overlap heavily: build_network, build_sop_chain, create_node, create_lop_node, create_cop_node, create_chop_node, and the setup_*_sim helpers all create nodes/networks with conflicting 'preferred' guidance. Capture tools also collide (render_node_network vs capture_network_editor, render_viewport vs capture_screenshot), as do diagnostic tools like get_node_info, explain_node, verify_network, and get_node_errors_detailed. The rich descriptions help, but the boundaries remain unclear in too many places for reliable selection.
The dominant convention is clear snake_case verb_noun (get_*, set_*, create_*, list_*, connect_*), which is readable and predictable. Minor inconsistencies appear: get_attrib_values vs get_attribute_info, new_scene instead of create_scene, playbar_control as noun_verb, render_node_network vs capture_network_editor for similar actions, and create_vex_expression actually sets an HScript expression. These are deviations rather than a breakdown of the naming scheme.
216 tools is far beyond any reasonable agent-facing surface, even for a domain as large as Houdini. The count forces heavy context use and makes tool selection, prioritization, and maintenance difficult, with many convenience wrappers layered over generic tools. This is an extreme mismatch relative to the 3-15 well-scoped ideal.
Coverage is unusually broad: node lifecycle, parameters/keyframes/expressions, SOP geometry, USD/LOP inspection, DOPs, TOPs, COPs, CHOPs, caches, rendering, viewport, HDAs, shelf tools, takes, and scene management are all represented. Gaps remain around direct geometry attribute writing, DOP field writes, node comment editing, and USD prim editing beyond attributes, but most have node/VEX/Python workarounds.