How do I install the Loonghao Dcc MCP Houdini Houdini Dev skill?

Run npx skillmds@latest add tomevault-io/loonghao-dcc-mcp-houdini-houdini-dev in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Loonghao Dcc MCP Houdini Houdini Dev skill do?

houdini-dev It is listed under AI & ML on SkillMD.

Is Loonghao Dcc MCP Houdini Houdini Dev safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Loonghao Dcc MCP Houdini Houdini Dev?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Loonghao Dcc MCP Houdini Houdini Dev free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Loonghao Dcc MCP Houdini Houdini Dev?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Loonghao Dcc MCP Houdini Houdini Dev by tomevault-io

Name: Loonghao Dcc MCP Houdini Houdini Dev
Author: tomevault-io

File contents

houdini-dev

Development workflow helpers for agents iterating on Houdini Python tools. Stays outside the minimal startup surface (load explicitly with load_skill("houdini-dev")). The intended loop:

attach_project("/path/to/tool") once for the project root.
Edit files in the normal workspace.
reload_modules(prefix="mytool") → run_entrypoint("mytool.cli:main").
Inspect the concise stdout/stderr/traceback summary.

Set DCC_MCP_HOUDINI_DEV_ROOTS to a path-list of trusted roots to restrict which local projects attach_project / run_script will accept.

Tool groups

dev-project: attach_project, reload_modules.
dev-run: run_entrypoint, run_script (captured output).
dev-debug: start_debugpy (safe 127.0.0.1:5678 default, already-running detection).
dev-introspect (read-only): introspect_hom — members/signatures, node-type categories, and available node types.
dev-ui: ui_snapshot (read-only), ui_action.

Context limitations

Headless hython: ui_snapshot / ui_action return a clean supported=false result instead of failing when no UI is available.
Trusted roots: when DCC_MCP_HOUDINI_DEV_ROOTS is set, project and script paths outside those roots are refused with a structured error.
debugpy: requires debugpy installed in the Houdini Python env; a second call reports already_running=true rather than re-binding the port.
HOM only with hou: introspect_hom and the UI tools require hou; they degrade through skill_error when it cannot be imported.

Tracer-bullet flow

attach_project("/projects/mytool")
introspect_hom(category="Sop", name_filter="attrib") → discover node types
reload_modules(prefix="mytool", reimport=true)
run_entrypoint("mytool.build:run", kwargs={"dry_run": true})
ui_snapshot() → confirm the active desktop when running interactively

Source: loonghao/dcc-mcp-houdini — distributed by TomeVault.

---name: loonghao-dcc-mcp-houdini-houdini-devdescription: houdini-dev---# houdini-devDevelopment workflow helpers for agents iterating on Houdini Python tools.Stays **outside** the minimal startup surface (load explicitly with`load_skill("houdini-dev")`). The intended loop:1. `attach_project("/path/to/tool")` once for the project root.2. Edit files in the normal workspace.3. `reload_modules(prefix="mytool")` → `run_entrypoint("mytool.cli:main")`.4. Inspect the concise stdout/stderr/traceback summary.Set `DCC_MCP_HOUDINI_DEV_ROOTS` to a path-list of trusted roots to restrictwhich local projects `attach_project` / `run_script` will accept.## Tool groups- **`dev-project`:** `attach_project`, `reload_modules`.- **`dev-run`:** `run_entrypoint`, `run_script` (captured output).- **`dev-debug`:** `start_debugpy` (safe `127.0.0.1:5678` default, already-running detection).- **`dev-introspect`** (read-only): `introspect_hom` — members/signatures, node-type categories, and available node types.- **`dev-ui`:** `ui_snapshot` (read-only), `ui_action`.## Context limitations- **Headless hython:** `ui_snapshot` / `ui_action` return a clean `supported=false` result instead of failing when no UI is available.- **Trusted roots:** when `DCC_MCP_HOUDINI_DEV_ROOTS` is set, project and script paths outside those roots are refused with a structured error.- **debugpy:** requires `debugpy` installed in the Houdini Python env; a second call reports `already_running=true` rather than re-binding the port.- **HOM only with hou:** `introspect_hom` and the UI tools require `hou`; they degrade through `skill_error` when it cannot be imported.## Tracer-bullet flow1. `attach_project("/projects/mytool")`2. `introspect_hom(category="Sop", name_filter="attrib")` → discover node types3. `reload_modules(prefix="mytool", reimport=true)`4. `run_entrypoint("mytool.build:run", kwargs={"dry_run": true})`5. `ui_snapshot()` → confirm the active desktop when running interactively---> Source: [loonghao/dcc-mcp-houdini](https://github.com/loonghao/dcc-mcp-houdini) — distributed by [TomeVault](https://tomevault.io).

Frequently asked questions

Loonghao Dcc MCP Houdini Houdini Dev