Roblox Studio workflow
Work as a collaborator inside an existing place, not as if it were a blank text repository. Targets
current Roblox Studio and rolling platform APIs.
When to use
- Use before invasive edits to a Roblox place, Studio DataModel, or Rojo-backed project.
- Use to plan execution location, find/reuse systems, create editable Instances, and gather honest
Studio verification evidence.
When not to use: this skill owns project operation, not the domain implementation. Compose it
with roblox-ui, roblox-networking, roblox-characters, roblox-physics, roblox-luau, or
roblox-datastores as appropriate.
Workflow
- Establish source of truth. Determine whether Studio, Rojo files, a package manager, or a
generated build owns each subtree. Do not hand-edit generated output or create Studio-only
changes that the next sync overwrites.
- Inventory before proposing structure. Inspect Explorer, search scripts/modules/remotes/tags,
read project mapping and naming conventions, and run once to observe existing warnings. Locate
the feature's current owner and dependencies.
- Plan the smallest compatible edit. Name the Instances and execution locations you will
change. Reuse existing remotes, component modules, cleanup utilities, tags, Attributes, folders,
and configuration patterns. State any unresolved ownership boundary.
- Create real, meaningful objects. Normal authored UI, folders, attachments, constraints,
remotes, and configuration remain inspectable in Explorer. Use scripts for behavior and truly
dynamic repetition—not to hide the entire feature in runtime construction.
- Respect execution and replication. Put authoritative logic in server containers, persistent
client logic in
StarterPlayerScripts, per-character client logic in
StarterCharacterScripts, UI templates in StarterGui, shared assets/modules/remotes in
ReplicatedStorage, and secrets/server assets in server-only containers.
- Test the actual risk. Choose Test, Test Here, Run, Server & Clients, Device Simulator,
Controller Emulator, Network Simulator, or scripted Studio testing based on the behavior.
Inspect server and client Output, not only the viewport.
- Clean the experiment. Remove temporary parts, test remotes, command-bar scripts, debug UI,
prints, disabled duplicate scripts, placeholder assets, and tags/Attributes created only for
diagnosis. Stop the session and confirm edit-mode hierarchy is clean.
- Report evidence precisely. List changed Instances/files, modes and client counts, cases
executed, observed Output, failures and fixes, and anything not run. Never translate “looks
plausible” or static inspection into “tested in Studio.”
Execution-location map
| Location |
Typical role |
Key constraint |
ServerScriptService |
authoritative server Scripts/modules |
not replicated to clients |
ServerStorage |
server-only assets/data |
clients cannot access it |
ReplicatedStorage |
shared modules/assets/remotes |
visible to clients; do not store secrets |
StarterPlayerScripts |
client systems across respawns |
clone into PlayerScripts |
StarterCharacterScripts |
per-character client behavior |
recreated each character |
StarterGui |
authored UI templates |
clones into PlayerGui; lifecycle depends on GUI policy |
Workspace |
replicated world and live characters |
streaming/ownership may affect client view |
Verify actual Script RunContext and project conventions; class/location shorthand does not excuse
inspection of an existing custom setup.
Reuse before adding
Before creating a RemoteEvent, search by purpose and inspect both endpoints. Before adding a
“manager” or service module, identify the current feature owner. Before introducing folders/tags,
inspect naming and retrieval patterns. Duplicate systems are especially dangerous when both run:
they double-bind input, save twice, create two UI clones, or apply gameplay twice.
Use Attributes for small typed designer-editable metadata that belongs on an Instance. Use
CollectionService tags for discovering sets of Instances across hierarchy. Use folders for
ownership/navigation, not as a substitute for semantic tags. Preserve established schemas and
validate missing/malformed Attributes at system boundaries.
local CollectionService = game:GetService("CollectionService")
local function setupDoor(instance: Instance)
if not instance:IsA("Model") then return end
local accessLevel = instance:GetAttribute("AccessLevel")
if type(accessLevel) ~= "number" then
warn(`Door {instance:GetFullName()} needs numeric AccessLevel`)
return
end
attachDoorBehavior(instance, accessLevel)
end
for _, door in CollectionService:GetTagged("Door") do setupDoor(door) end
CollectionService:GetInstanceAddedSignal("Door"):Connect(setupDoor)
The owning system must also clean behavior when a tagged Instance is removed/destroyed.
Studio MCP server (optional agent path)
Roblox Studio has a built-in MCP server. When it is connected to your AI client, drive the
same inspect -> edit -> run/verify -> playtest ladder through its tools instead of manual
clicking. This is OPTIONAL: when it is not connected, the manual Studio + Rojo workflow
above and below is the full default path, and nothing in this skill requires MCP, Assistant,
or any other tool. Enable it in Studio under Assistant, then Manage MCP Servers, then
"Enable Studio as MCP server" (it speaks MCP over stdio to the client).
Map the workflow onto the documented tools. Every call also takes a studio_id; use
list_roblox_studios to pick the target when several Studio windows are open.
| Ladder step |
MCP tools |
| inspect / inventory |
search_game_tree, inspect_instance, script_search, script_grep |
| edit |
script_read, multi_edit (creates the script when the path is new) |
| run / verify |
execute_luau (needs a datamodel_type: Edit, Client, or Server), get_studio_state, start_stop_play, get_console_output, screen_capture |
| playtest |
subagent (types explore, playtest), character_navigation, user_keyboard_input, user_mouse_input |
| assets |
search_asset, insert_asset, generate_mesh/generate_material/generate_procedural_model (+ wait_job_finished) |
| Roblox's own docs/skills |
http_get (allowed Roblox doc URLs), skill (Roblox-authored reference) |
The discipline does not change with MCP: multi_edit still writes normal, inspectable Scripts
(not a feature hidden in runtime construction), the server stays authoritative, you still edit
source-of-truth files rather than generated output, and you still clean up experiments. Treat
get_console_output and screen_capture as the agent-native evidence for the verification
ladder, and still inspect BOTH server and client output. An inserted Creator Store model can
carry its own Scripts: inspect its descendants before keeping it, as you would any third-party
code. start_stop_play runs a single solo session; multi-client checks still need Server &
Clients or StudioTestService (see the verification reference). Connected clients can read and
modify every open place, so connect only trusted ones. MCP is a transport for the same workflow,
not a licence to skip it.
Roblox also ships agent-readable docs independent of MCP: an index at
https://create.roblox.com/docs/llms.txt and a .md twin of every docs page (append .md
to its URL). A connected client can reach these through http_get/skill; an offline agent
reads the .md pages directly.
Verification ladder
- Static: project mapping/JSON/XML parses; links resolve; Luau diagnostics/lint if available.
- Single session: correct execution context, hierarchy, no new Output errors/warnings, basic
happy path and cleanup.
- Lifecycle: respawn, reset, reopen/rejoin, streaming or character replacement as relevant.
- Server & Clients: at least two clients for remotes, replication, ownership, player
interaction, joins/leaves, and targeted state.
- Device/input: representative device profiles, orientation, touch, controller navigation.
- Stress/failure: malformed input, spam, empty/full content, rapid transitions, simultaneous
actions, disconnects, and cleanup.
Use the smallest rungs that cover the change, but never substitute a lower rung for a claimed
higher-rung result.
Common failures
| Symptom |
Likely cause |
Remedy |
| Studio edit disappears after sync |
generated/mapped subtree edited |
change source-of-truth files and resync |
| feature runs twice |
duplicate Script/remote/controller |
inspect first; consolidate under existing owner |
| client cannot access module/object |
wrong execution/storage location |
move shared content deliberately; keep secrets server-only |
| authored screen unreadable in Explorer |
everything created by one LocalScript |
create named Instance shell and small behavior modules |
| test passes only in Test (solo) |
replication/lifecycle untested |
use Server & Clients and relevant emulators |
| “clean” Output hides client errors |
only server Output inspected |
inspect each client and server context |
| shipped place has DebugFolder/prints |
experiment cleanup skipped |
maintain cleanup list and inspect hierarchy after session |
MCP tools load but list_roblox_studios is empty |
Studio closed, no place open, or MCP setting off |
open a place and enable "Enable Studio as MCP server" |
| MCP edit lands in the wrong place |
several Studio windows open |
pick the studio_id from list_roblox_studios by place name |
| Attribute/tag behavior silently fails |
schema/class not validated |
validate at setup boundary; warn with full Instance path |
Resources
- Read
references/studio-verification.md before reporting Studio validation or when choosing test
modes, constructing a temporary harness, or documenting an automation gap.
Related skills
roblox-luau — services, Instances, events, and basic execution model.
roblox-ui, roblox-networking, roblox-characters, roblox-physics — focused production
workflows that use this inspection and verification discipline.
Primary references
https://create.roblox.com/docs/studio/testing-modes
https://create.roblox.com/docs/projects/data-model
https://create.roblox.com/docs/scripting/services
https://create.roblox.com/docs/studio/mcp
1---2name: roblox-studio-workflow3description: >4 Modify living Roblox Studio projects safely: inspect Explorer and execution locations, preserve5 existing structure, reuse modules/remotes, create meaningful Instances, apply Attributes and6 CollectionService tags, test server/client and multiple devices, inspect Output, and remove7 experiments/debug objects. Use when an agent edits a place or rbxl/rbxlx/Rojo project, operates8 in Studio, adds Scripts/LocalScripts/ModuleScripts, or must verify a Roblox change in context.9metadata:10 source: https://skillmd.com/skills/junmystery/roblox-studio-workflow11---1213# Roblox Studio workflow1415Work as a collaborator inside an existing place, not as if it were a blank text repository. Targets16current Roblox Studio and rolling platform APIs.1718## When to use1920- Use before invasive edits to a Roblox place, Studio DataModel, or Rojo-backed project.21- Use to plan execution location, find/reuse systems, create editable Instances, and gather honest22 Studio verification evidence.2324**When not to use:** this skill owns project operation, not the domain implementation. Compose it25with `roblox-ui`, `roblox-networking`, `roblox-characters`, `roblox-physics`, `roblox-luau`, or26`roblox-datastores` as appropriate.2728## Workflow29301. **Establish source of truth.** Determine whether Studio, Rojo files, a package manager, or a31 generated build owns each subtree. Do not hand-edit generated output or create Studio-only32 changes that the next sync overwrites.332. **Inventory before proposing structure.** Inspect Explorer, search scripts/modules/remotes/tags,34 read project mapping and naming conventions, and run once to observe existing warnings. Locate35 the feature's current owner and dependencies.363. **Plan the smallest compatible edit.** Name the Instances and execution locations you will37 change. Reuse existing remotes, component modules, cleanup utilities, tags, Attributes, folders,38 and configuration patterns. State any unresolved ownership boundary.394. **Create real, meaningful objects.** Normal authored UI, folders, attachments, constraints,40 remotes, and configuration remain inspectable in Explorer. Use scripts for behavior and truly41 dynamic repetition—not to hide the entire feature in runtime construction.425. **Respect execution and replication.** Put authoritative logic in server containers, persistent43 client logic in `StarterPlayerScripts`, per-character client logic in44 `StarterCharacterScripts`, UI templates in `StarterGui`, shared assets/modules/remotes in45 `ReplicatedStorage`, and secrets/server assets in server-only containers.466. **Test the actual risk.** Choose Test, Test Here, Run, Server & Clients, Device Simulator,47 Controller Emulator, Network Simulator, or scripted Studio testing based on the behavior.48 Inspect server and client Output, not only the viewport.497. **Clean the experiment.** Remove temporary parts, test remotes, command-bar scripts, debug UI,50 prints, disabled duplicate scripts, placeholder assets, and tags/Attributes created only for51 diagnosis. Stop the session and confirm edit-mode hierarchy is clean.528. **Report evidence precisely.** List changed Instances/files, modes and client counts, cases53 executed, observed Output, failures and fixes, and anything not run. Never translate “looks54 plausible” or static inspection into “tested in Studio.”5556## Execution-location map5758| Location | Typical role | Key constraint |59|---|---|---|60| `ServerScriptService` | authoritative server Scripts/modules | not replicated to clients |61| `ServerStorage` | server-only assets/data | clients cannot access it |62| `ReplicatedStorage` | shared modules/assets/remotes | visible to clients; do not store secrets |63| `StarterPlayerScripts` | client systems across respawns | clone into `PlayerScripts` |64| `StarterCharacterScripts` | per-character client behavior | recreated each character |65| `StarterGui` | authored UI templates | clones into `PlayerGui`; lifecycle depends on GUI policy |66| `Workspace` | replicated world and live characters | streaming/ownership may affect client view |6768Verify actual Script `RunContext` and project conventions; class/location shorthand does not excuse69inspection of an existing custom setup.7071## Reuse before adding7273Before creating a RemoteEvent, search by purpose and inspect both endpoints. Before adding a74“manager” or service module, identify the current feature owner. Before introducing folders/tags,75inspect naming and retrieval patterns. Duplicate systems are especially dangerous when both run:76they double-bind input, save twice, create two UI clones, or apply gameplay twice.7778Use Attributes for small typed designer-editable metadata that belongs on an Instance. Use79`CollectionService` tags for discovering sets of Instances across hierarchy. Use folders for80ownership/navigation, not as a substitute for semantic tags. Preserve established schemas and81validate missing/malformed Attributes at system boundaries.8283```lua84local CollectionService = game:GetService("CollectionService")8586local function setupDoor(instance: Instance)87 if not instance:IsA("Model") then return end88 local accessLevel = instance:GetAttribute("AccessLevel")89 if type(accessLevel) ~= "number" then90 warn(`Door {instance:GetFullName()} needs numeric AccessLevel`)91 return92 end93 attachDoorBehavior(instance, accessLevel)94end9596for _, door in CollectionService:GetTagged("Door") do setupDoor(door) end97CollectionService:GetInstanceAddedSignal("Door"):Connect(setupDoor)98```99100The owning system must also clean behavior when a tagged Instance is removed/destroyed.101102## Studio MCP server (optional agent path)103104Roblox Studio has a built-in MCP server. When it is connected to your AI client, drive the105same inspect -> edit -> run/verify -> playtest ladder through its tools instead of manual106clicking. This is OPTIONAL: when it is not connected, the manual Studio + Rojo workflow107above and below is the full default path, and nothing in this skill requires MCP, Assistant,108or any other tool. Enable it in Studio under Assistant, then Manage MCP Servers, then109"Enable Studio as MCP server" (it speaks MCP over stdio to the client).110111Map the workflow onto the documented tools. Every call also takes a `studio_id`; use112`list_roblox_studios` to pick the target when several Studio windows are open.113114| Ladder step | MCP tools |115|---|---|116| inspect / inventory | `search_game_tree`, `inspect_instance`, `script_search`, `script_grep` |117| edit | `script_read`, `multi_edit` (creates the script when the path is new) |118| run / verify | `execute_luau` (needs a `datamodel_type`: Edit, Client, or Server), `get_studio_state`, `start_stop_play`, `get_console_output`, `screen_capture` |119| playtest | `subagent` (types `explore`, `playtest`), `character_navigation`, `user_keyboard_input`, `user_mouse_input` |120| assets | `search_asset`, `insert_asset`, `generate_mesh`/`generate_material`/`generate_procedural_model` (+ `wait_job_finished`) |121| Roblox's own docs/skills | `http_get` (allowed Roblox doc URLs), `skill` (Roblox-authored reference) |122123The discipline does not change with MCP: `multi_edit` still writes normal, inspectable Scripts124(not a feature hidden in runtime construction), the server stays authoritative, you still edit125source-of-truth files rather than generated output, and you still clean up experiments. Treat126`get_console_output` and `screen_capture` as the agent-native evidence for the verification127ladder, and still inspect BOTH server and client output. An inserted Creator Store model can128carry its own Scripts: inspect its descendants before keeping it, as you would any third-party129code. `start_stop_play` runs a single solo session; multi-client checks still need Server &130Clients or `StudioTestService` (see the verification reference). Connected clients can read and131modify every open place, so connect only trusted ones. MCP is a transport for the same workflow,132not a licence to skip it.133134Roblox also ships agent-readable docs independent of MCP: an index at135`https://create.roblox.com/docs/llms.txt` and a `.md` twin of every docs page (append `.md`136to its URL). A connected client can reach these through `http_get`/`skill`; an offline agent137reads the `.md` pages directly.138139## Verification ladder1401411. **Static:** project mapping/JSON/XML parses; links resolve; Luau diagnostics/lint if available.1422. **Single session:** correct execution context, hierarchy, no new Output errors/warnings, basic143 happy path and cleanup.1443. **Lifecycle:** respawn, reset, reopen/rejoin, streaming or character replacement as relevant.1454. **Server & Clients:** at least two clients for remotes, replication, ownership, player146 interaction, joins/leaves, and targeted state.1475. **Device/input:** representative device profiles, orientation, touch, controller navigation.1486. **Stress/failure:** malformed input, spam, empty/full content, rapid transitions, simultaneous149 actions, disconnects, and cleanup.150151Use the smallest rungs that cover the change, but never substitute a lower rung for a claimed152higher-rung result.153154## Common failures155156| Symptom | Likely cause | Remedy |157|---|---|---|158| Studio edit disappears after sync | generated/mapped subtree edited | change source-of-truth files and resync |159| feature runs twice | duplicate Script/remote/controller | inspect first; consolidate under existing owner |160| client cannot access module/object | wrong execution/storage location | move shared content deliberately; keep secrets server-only |161| authored screen unreadable in Explorer | everything created by one LocalScript | create named Instance shell and small behavior modules |162| test passes only in Test (solo) | replication/lifecycle untested | use Server & Clients and relevant emulators |163| “clean” Output hides client errors | only server Output inspected | inspect each client and server context |164| shipped place has DebugFolder/prints | experiment cleanup skipped | maintain cleanup list and inspect hierarchy after session |165| MCP tools load but `list_roblox_studios` is empty | Studio closed, no place open, or MCP setting off | open a place and enable "Enable Studio as MCP server" |166| MCP edit lands in the wrong place | several Studio windows open | pick the `studio_id` from `list_roblox_studios` by place name |167| Attribute/tag behavior silently fails | schema/class not validated | validate at setup boundary; warn with full Instance path |168169## Resources170171- Read `references/studio-verification.md` before reporting Studio validation or when choosing test172 modes, constructing a temporary harness, or documenting an automation gap.173174## Related skills175176- `roblox-luau` — services, Instances, events, and basic execution model.177- `roblox-ui`, `roblox-networking`, `roblox-characters`, `roblox-physics` — focused production178 workflows that use this inspection and verification discipline.179180## Primary references181182- `https://create.roblox.com/docs/studio/testing-modes`183- `https://create.roblox.com/docs/projects/data-model`184- `https://create.roblox.com/docs/scripting/services`185- `https://create.roblox.com/docs/studio/mcp`