$npx -y skills add TabooHarmony/roblox-brain --skill roblox-studio-mcpUse when working with Roblox Studio through built-in MCP for scripts, scenes, generated assets, input, or playtesting.
| 1 | # Roblox Studio MCP |
| 2 | |
| 3 | ## When to Load |
| 4 | |
| 5 | Load when working with Roblox Studio through its MCP server: inspecting or editing scripts, building maps and props, generating or inserting assets, debugging, or playtesting. Skip for standalone code generation with no Studio connection. |
| 6 | |
| 7 | ## Quick Reference |
| 8 | |
| 9 | ### Bootstrap before mutation |
| 10 | 1. `list_roblox_studios` and identify the target. |
| 11 | 2. `set_active_studio` with the returned ID when selection matters. |
| 12 | 3. `get_studio_state` and confirm Edit/Client/Server availability. |
| 13 | 4. Inspect the target tree and scripts before changing them. |
| 14 | |
| 15 | Use the actual exposed schema. Pass `datamodel_type` only where the tool requires it: `Edit` for edit-time changes, `Client` or `Server` for runtime operations. Do not guess from a previous session. |
| 16 | |
| 17 | ### Capabilities |
| 18 | - **Inspect:** `search_game_tree`, `inspect_instance`, `script_search`, `script_read`, `script_grep` |
| 19 | - **Edit/execute:** `multi_edit`, `execute_luau` |
| 20 | - **Assets:** search and insert existing assets; `generate_mesh`, `generate_material`, `generate_procedural_model`, `wait_job_finished`, `store_image`, `upload_image` |
| 21 | - **Play/evidence:** `start_stop_play`, `get_console_output`, `screen_capture`, `character_navigation`, `user_keyboard_input`, `user_mouse_input`, `subagent` |
| 22 | |
| 23 | The docs use names such as `search_asset` and `insert_asset`; other bridges may expose aliases such as `search_creator_store` and `insert_from_creator_store`. Route by capability, then inspect the live schema. |
| 24 | |
| 25 | ### Execution contract |
| 26 | ```text |
| 27 | discover → select Studio/context → inspect → mutate in bounded batches |
| 28 | → read back → playtest when needed → inspect console/visual evidence |
| 29 | → clean up or report the exact fallback |
| 30 | ``` |
| 31 | |
| 32 | ### Reliability rules |
| 33 | - `execute_luau` is stateless. Re-acquire references every call. |
| 34 | - Read before write and read back after every script, asset, or geometry mutation. |
| 35 | - Generate or insert assets only after choosing between reuse, procedural generation, mesh/material generation, and native fallback. |
| 36 | - Treat generation as asynchronous when a generation ID is returned; call `wait_job_finished` with that `generationId` before dependent edits. |
| 37 | - Keep large scripts in `multi_edit` chunks, not one oversized execution payload. |
| 38 | - If MCP is absent or a capability is unavailable, provide complete offline Luau and state what was not verified. |
| 39 | |
| 40 | > Full tool mappings, live-schema differences, asset workflows, evidence recipes, and recovery rules: [references/full.md](references/full.md) |