> ## Documentation Index > Fetch the complete documentation index at: https://docs.arcprize.org/llms.txt > Use this file to discover all available pages before exploring further. # Execute simple action 7 > Issues **ACTION 7** to the specified game session. This is a single-parameter command (no X/Y coordinates): ACTION7 will always be an undo action for games that support it. The request must include: • `game_id` — which game to act on • `guid` — the active session identifier returned from RESET An optional `reasoning` JSON blob (≤ 16 KB) can be attached for audit or research purposes. A successful call returns the next visual frame(s) and updated score/state. ## OpenAPI ````yaml /arc3v1.yaml post /api/cmd/ACTION7 openapi: 3.0.3 info: title: ARC‑AGI‑3 REST API version: 1.0.0 description: > Programmatic interface for running agents against ARC‑AGI‑3 games, opening/closing score‑cards and driving game state with actions. All requests **require** an `X‑API‑Key` header issued from the ARC‑AGI‑3 web console. **Important: Session Affinity via Cookies** Games are stateful and require session affinity. The server sets cookies (especially `AWSALB*` cookies) in responses that **must be included in all subsequent requests** for the same game session. These cookies route requests to the correct backend instance maintaining your game state. Most HTTP clients handle cookies automatically, but ensure your client preserves and sends cookies received from RESET and ACTION responses. servers: - url: https://three.arcprize.org security: [] tags: - name: Games - name: Scorecards - name: Commands paths: /api/cmd/ACTION7: post: tags: - Commands summary: Execute simple action 7 description: | Issues **ACTION 7** to the specified game session. This is a single-parameter command (no X/Y coordinates): ACTION7 will always be an undo action for games that support it. The request must include: • `game_id` — which game to act on • `guid` — the active session identifier returned from RESET An optional `reasoning` JSON blob (≤ 16 KB) can be attached for audit or research purposes. A successful call returns the next visual frame(s) and updated score/state. operationId: action7 requestBody: description: Game/session identifiers plus optional reasoning data. required: true content: application/json: schema: $ref: '#/components/schemas/SimpleActionCommand' examples: action: value: game_id: ls20-016295f7601e guid: 2fa5332c-2e55-4825-b5c5-df960d504470 reasoning: policy: π_left responses: '200': description: Frame returned after executing the action. content: application/json: schema: $ref: '#/components/schemas/FrameResponse' examples: frame: value: game_id: ls20-016295f7601e guid: 2fa5332c-2e55-4825-b5c5-df960d504470 frame: - - - 0 - 0 - 1 - … - - … state: NOT_FINISHED levels_completed: 3 win_levels: 254 action_input: id: 7 available_actions: - 1 - 2 - 3 - 4 - 7 '400': description: | Bad request - possible causes: • Unknown `game_id` or invalid format • `guid` not found or does not belong to `game_id` • `reasoning` field exceeds 16 KB or is malformed '401': description: Missing or invalid **X-API-Key** header. security: - ApiKeyAuth: [] components: schemas: SimpleActionCommand: type: object description: > Issues a one-parameter action (ACTION1 - ACTION5) to a running game instance identified by `guid`. **Important:** Include any cookies (especially `AWSALB*` cookies) received from previous RESET or ACTION responses to ensure session affinity. properties: game_id: type: string description: Game identifier this action targets. guid: type: string description: Server-generated session ID obtained from a RESET response. reasoning: type: object description: | Optional, caller-defined JSON blob (≤ 16 KB) capturing the agent's internal reasoning, model parameters, or any other metadata you'd like to store alongside the action. additionalProperties: true required: - game_id - guid FrameResponse: type: object description: | Snapshot returned after every RESET or ACTION command. Includes the latest visual frame(s), cumulative score details, the current game state, and an echo of the triggering action. properties: game_id: type: string description: Game identifier for the running session. guid: type: string description: Server-generated session ID; use this for all subsequent commands. frame: type: array description: | One or more consecutive visual frames. Each frame is a 64 × 64 grid of 4-bit colour indices (integers 0-15). Multiple frames may be returned if the environment advances internally (e.g., animations) before settling. items: type: array items: type: array items: type: integer minimum: 0 maximum: 15 state: type: string description: > Current state of the session: • **NOT_FINISHED** - game in progress, not yet WIN or GAME_OVER. • **NOT_STARTED** - session has ended (WIN or GAME_OVER) and requires RESET. • **WIN** - session ended in victory. • **GAME_OVER** - session ended in defeat. enum: - NOT_FINISHED - NOT_STARTED - WIN - GAME_OVER levels_completed: type: integer description: Current cumulative number of levels completed for this run. minimum: 0 maximum: 254 win_levels: type: integer description: | Level threshold required to reach the **WIN** state. Mirrors the game's configured win condition so agents can adapt dynamically without hard-coding values. minimum: 0 maximum: 254 action_input: type: object description: Echo of the command that produced this frame. properties: id: type: integer description: Client-assigned or sequential action index. data: type: object description: Additional parameters originally sent with the action. additionalProperties: true available_actions: type: array description: List of available actions for the current game. items: type: integer enum: - 1 - 2 - 3 - 4 - 5 - 6 required: - game_id - guid - frame - state - levels_completed - win_levels - action_input - available_actions securitySchemes: ApiKeyAuth: type: apiKey in: header name: X-API-Key ````