Skip to main content
Glama
EL4CTEO

Roblox Studio MCP

Drive the player during a playtest

character

Moves and acts as the player in a running playtest to test gameplay without user input. Walks to positions, performs actions like jump or use tool, and reports state.

Instructions

Moves and acts as the player character in a running playtest, so gameplay can be tested without asking the user to play it.

moveTo walks to a position or to an instance, following a path computed around walls and gaps rather than a straight line into them. It reports whether it ACTUALLY ARRIVED and how far short it stopped — a route blocked by something you did not know about otherwise looks identical to a successful walk.

act does the one-shot things worth testing: jump, sit, stand, respawn, kill (to exercise the death and respawn path), teleport, and equip/activate to use a Tool — which is how combat gets tested, since Activate is exactly what a mouse click triggers. Note that teleport skips everything in between, so triggers and collisions along the route do not fire — walk if you are testing those.

state reports position, health, walk speed and what the humanoid is doing. Call it before and after anything else here.

This drives the Humanoid directly rather than simulating keystrokes, which is the right tool for going places: pathfinding around a wall is one call here and a sequence of guessed key presses otherwise. For anything bound to a control rather than to movement — does E open the door, does the sprint key work, does Escape close the menu — use input, which sends real key and mouse events.

REQUIRES A RUNNING PLAYTEST, and the character lives in the playtest's data model — address these to the playtest's studioId from list_studios, not the editor's. Run mode has no character at all; use playtest op=play.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
opYes'moveTo' walks somewhere, 'path' checks a route without walking it, 'act' performs an action, 'state' only reports.
toNoTarget position, e.g. "25, 5, -10". Used by moveTo and by teleport.
fromNopath only: where the route starts, e.g. "0, 5, 0". Defaults to the character.
pathNomoveTo only: walk to this instance instead of a coordinate.
toolNoequip only: the Tool's name.
costsNoMaterial or PathfindingModifier label → cost, e.g. { "Water": 20 } to avoid swimming. Higher is more avoided; the route chosen is the cheapest total, not the shortest.
actionNoact only: what to do. 'equip' takes a Tool from the Backpack or StarterPack, 'activate' uses it (what a mouse click triggers).
directNomoveTo only: walk straight at the target without pathfinding. Use when a route is reported unreachable but you want to see what happens.
playerNoWhich player, by name. Omit for the only one; needed in a multiplayer test.
toPathNopath only: an instance to end at instead of `to`.
canJumpNomoveTo only: allow the path to include jumps.
spacingNoStuds between waypoints, default 4. Tighter follows the geometry more closely; wider is a coarser route.
canClimbNoWhether it may climb truss. Off by default.
fromPathNopath only: an instance to start from, e.g. "Workspace.SpawnLocation".
studioIdNoThe PLAYTEST session's id — not the editor's. See list_studios.
agentHeightNoHow tall the walker is. Defaults to 5, a standard character.
agentRadiusNoHow wide the walker is, in studs. Defaults to 2 — a standard character. Raise it to ask whether a bigger NPC fits through the same gaps a player does.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed10 schema fields changedv0.6.5
    • addedInput schema / properties / agentHeight
      Added value: +{
      +  "description": "How tall the walker is. Defaults to 5, a standard character.",
      +  "maximum": 100,
      +  "minimum": 0.1,
      +  "type": "number"
      +}
    • addedInput schema / properties / agentRadius
      Added value: +{
      +  "description": "How wide the walker is, in studs. Defaults to 2 — a standard character. Raise it to ask whether a bigger NPC fits through the same gaps a player does.",
      +  "maximum": 50,
      +  "minimum": 0.1,
      +  "type": "number"
      +}
    • addedInput schema / properties / canClimb
      Added value: +{
      +  "description": "Whether it may climb truss. Off by default.",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / costs
      Added value: +{
      +  "additionalProperties": {
      +    "type": "number"
      +  },
      +  "description": "Material or PathfindingModifier label → cost, e.g. { \"Water\": 20 } to avoid swimming. Higher is more avoided; the route chosen is the cheapest total, not the shortest.",
      +  "propertyNames": {
      +    "type": "string"
      +  },
      +  "type": "object"
      +}
    • addedInput schema / properties / from
      Added value: +{
      +  "description": "path only: where the route starts, e.g. \"0, 5, 0\". Defaults to the character.",
      +  "type": "string"
      +}
    • addedInput schema / properties / fromPath
      Added value: +{
      +  "description": "path only: an instance to start from, e.g. \"Workspace.SpawnLocation\".",
      +  "type": "string"
      +}
    • changedInput schema / properties / op / description
      Previous value: -"'moveTo' walks somewhere, 'act' performs an action, 'state' only reports."New value: +"'moveTo' walks somewhere, 'path' checks a route without walking it, 'act' performs an action, 'state' only reports."
    • changedInput schema / properties / op / enum
      Previous value: -[
      -  "moveTo",
      -  "act",
      -  "state"
      -]New value: +[
      +  "moveTo",
      +  "path",
      +  "act",
      +  "state"
      +]
    • addedInput schema / properties / spacing
      Added value: +{
      +  "description": "Studs between waypoints, default 4. Tighter follows the geometry more closely; wider is a coarser route.",
      +  "maximum": 100,
      +  "minimum": 0.1,
      +  "type": "number"
      +}
    • addedInput schema / properties / toPath
      Added value: +{
      +  "description": "path only: an instance to end at instead of `to`.",
      +  "type": "string"
      +}
  2. First observedv0.1.8

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare generic flags (readOnlyHint=false, openWorldHint=true, destructiveHint=false, idempotentHint=false). The description adds real behavioral context beyond them: `kill` exercises the death/respawn path, `teleport` skips triggers and collisions, `moveTo` reports whether it ACTUALLY ARRIVED and how far short, and a running playtest is required. This is materially more than the annotations convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core purpose, then well-paragraphed by op and by sibling routing; almost every sentence carries actionable detail. Slightly long and a few clauses are decorative ('which is how combat gets tested'), keeping it short of a 5.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 17-parameter, multi-mode tool with openWorld semantics and no output schema, the description covers the operational context an agent needs (preconditions, mode selection, sibling routing, key side-effect caveats). It stops short only on how `state`/`path` results are shaped, which the missing output schema would otherwise have to cover.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3 and the schema already carries the per-parameter documentation. The description nevertheless adds cross-parameter semantics that the schema alone doesn't: teleport-vs-walk trade-offs, the equip/activate chain as the combat-test path, and the `costs` avoidance intent. That is worth one step above baseline.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a concrete verb+resource (drives the player character during a running playtest) and enumerates its `op` modes with distinct purposes. It explicitly distinguishes itself from the `input` sibling for control-bound tests and from `playtest op=play` for entering run mode.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit routing: use `input` for anything bound to a control rather than movement, walk rather than teleport when testing triggers/collisions, call `state` before and after anything else, and note it requires a running playtest with the playtest's studioId, not the editor's.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.