像塑 MCP Plus
Related Servers
Alternatives to 像塑 MCP Plus
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to operate Douyin Xiangsu AR effects through MCP, including scene editing, asset binding, TypeScript scripting, compilation, preview screenshots, and interactive playtesting.20MIT
- AlicenseAqualityBmaintenanceConnects any MCP client to TikTok Effect House, letting AI agents build and edit effects live by conversation — running JavaScript cells against the editor's scene, script, asset and preview APIs, with docs search and editor launching. It also exposes the editor's automation tools for scene editing and preview captures, without needing the in-app AI panel.8MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients such as Claude Code, Cursor, and Codex to drive a live Cocos Creator 3.8.x editor and its project files, covering scene, node, component, asset, build, and view operations. Write operations run as undo-journaled transactions that roll back on failure, with a verified editor-message whitelist, path guards, tiered confirmations, and offline prefab/animation editing that works even when the editor is closed.MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI clients to control Cocos Creator editor projects, scenes, nodes, components, assets, Prefabs, building, and diagnostics via the MCP protocol.-
- FlicenseNot gradedqualityCmaintenanceEnables MCP clients such as Codex to drive a Cocos Creator 3.8.8 editor through four tools (status, catalog, execute, observe), covering scene/node/component editing, UI templates and adaptive layout, prefabs, script events, assets and dependency graphs, native animation and Spine, screenshots, and Web build verification. Operations run through a local extension bridge with batching, dry-run pre-checks, and read-back validation.-
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to directly control the Cocos Creator 3.8.x editor via MCP protocol, providing over 130 tools for scene, node, component, asset, and project operations.22 npm44MIT
TDQS
Scored across 49 tools
Each tool targets a distinct resource+action and the descriptions go to unusual lengths to state when to use one over another (library_list vs library_search, works_download vs works_open, menu_add vs library_apply, object_remove vs resource_delete_unused). The only real confusion points are the parallel browse/search pairs (library_*, works_*) and the two catalogs (library_list = 素材库 vs menu_list = 添加对象菜单), which an agent can resolve from the text but must read carefully.
All names are snake_case and the dominant convention is domain-prefixed noun_verb (works_list, library_apply, inspiration_bind, music_apply, menu_add, ai_combine_new, face_warp_set_intensity), which clusters nicely by domain. Deviations are minor but real: tasks_list breaks the singular task_* family, object_remove vs resource_delete_unused use different verbs for delete, and several tools are noun-only labels (scene_tree, makeup_layers, growth_route, my_effects, ext_status, image_view).
49 tools is a heavy surface for an agent to select from, and a few groups could plausibly collapse (three near-parallel list/search pairs, rank_list/subjects/innovation/hotspots, icon_set/icon_open). Still, the domain is genuinely enormous — editor scene inspection, material library, homepage works, inspirations, tasks, rankings, music, submission and preview — and most tools map to a distinct editor action, so this is borderline heavy rather than bloated.
Coverage spans the whole workflow: project open/new/close, material browse→apply, works browse→download→open→cleanup, inspiration/task binding, scene introspection (scene_tree, inspect_many, makeup_layers, face_warp_*), object/resource removal, icon, effect check, mobile preview and submit prep. Minor gaps remain — saving, undo, and capturing previews are delegated to the official editor tools rather than surfaced here, and there is no local asset import beyond images/icons.