Skip to main content
Glama

mgba_screenshot

Take a screenshot of the current GBA emulator display and save it as a PNG. Accepts an optional file path; returns the saved file path.

Instructions

Take a screenshot of the current display and save it to a file. Returns the saved file path.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pathNoAbsolute file path to save the PNG (optional — defaults to a temp file)

Implementation Reference

  • Schema/tool definition for mgba_screenshot — declares the tool name, description, and optional 'path' input parameter.
    {
      name: "mgba_screenshot",
      description: "Take a screenshot of the current display and save it to a file. Returns the saved file path.",
      inputSchema: {
        type: "object",
        properties: {
          path: {
            type: "string",
            description: "Absolute file path to save the PNG (optional — defaults to a temp file)",
          },
        },
      },
    },
  • Handler for mgba_screenshot — calls the mGBA bridge's 'screenshot' RPC method with an optional path, and returns the saved file path.
    case "mgba_screenshot": {
      const path = await mgba.call<string>("screenshot", p.path ? { path: p.path } : {});
      return ok(`Screenshot saved: ${path}`);
    }
  • src/tools.ts:258-411 (registration)
    Registration function that wires up both ListToolsRequestSchema (serving TOOLS array) and CallToolRequestSchema (routing the 'mgba_screenshot' case to the handler).
    export function registerTools(server: Server, mgba: MgbaClient): void {
      server.setRequestHandler(ListToolsRequestSchema, async () => ({ tools: TOOLS }));
    
      server.setRequestHandler(CallToolRequestSchema, async (req) => {
        const { name, arguments: args = {} } = req.params;
        const p = args as Record<string, unknown>;
    
        switch (name) {
          case "mgba_ping": {
            const r = await mgba.call<string>("ping");
            return ok(r);
          }
    
          case "mgba_get_info": {
            const r = await mgba.call<{
              title?: string;
              code?: string;
              frame?: number;
              platform?: number | string;
              capabilities?: Record<string, boolean>;
            }>("get_info");
            const lines = [
              `Title:    ${r.title ?? "(unavailable)"}`,
              `Code:     ${r.code ?? "(unavailable)"}`,
              `Platform: ${r.platform ?? "(unavailable)"}`,
              `Frame:    ${r.frame ?? "(unavailable)"}`,
            ];
            if (r.capabilities) {
              const present = Object.entries(r.capabilities).filter(([, v]) => v).map(([k]) => k);
              const missing = Object.entries(r.capabilities).filter(([, v]) => !v).map(([k]) => k);
              lines.push("");
              lines.push(`Capabilities present: ${present.length ? present.join(", ") : "(none)"}`);
              if (missing.length) lines.push(`Missing on this build: ${missing.join(", ")}`);
            }
            return ok(lines.join("\n"));
          }
    
          case "mgba_read8": {
            const v = await mgba.call<number>("read8", { address: p.address });
            return ok(`0x${(p.address as number).toString(16).toUpperCase()}: ${formatHex(v)}`);
          }
    
          case "mgba_read16": {
            const v = await mgba.call<number>("read16", { address: p.address });
            return ok(`0x${(p.address as number).toString(16).toUpperCase()}: ${formatHex(v)}`);
          }
    
          case "mgba_read32": {
            const v = await mgba.call<number>("read32", { address: p.address });
            return ok(`0x${(p.address as number).toString(16).toUpperCase()}: ${formatHex(v)}`);
          }
    
          case "mgba_write8": {
            await mgba.call("write8", { address: p.address, value: p.value });
            return ok(`Wrote ${formatHex(p.value)} → 0x${(p.address as number).toString(16).toUpperCase()}`);
          }
    
          case "mgba_write16": {
            await mgba.call("write16", { address: p.address, value: p.value });
            return ok(`Wrote ${formatHex(p.value)} → 0x${(p.address as number).toString(16).toUpperCase()}`);
          }
    
          case "mgba_write32": {
            await mgba.call("write32", { address: p.address, value: p.value });
            return ok(`Wrote ${formatHex(p.value)} → 0x${(p.address as number).toString(16).toUpperCase()}`);
          }
    
          case "mgba_read_range": {
            const bytes = await mgba.call<number[]>("read_range", {
              address: p.address,
              length:  p.length,
            });
            const hex = bytes
              .map((b) => b.toString(16).padStart(2, "0").toUpperCase())
              .join(" ");
            const addr = (p.address as number).toString(16).toUpperCase();
            return ok(`0x${addr} [${bytes.length} bytes]:\n${hex}`);
          }
    
          case "mgba_write_range": {
            const r = await mgba.call<{ written: number }>("write_range", {
              address: p.address,
              bytes:   p.bytes,
            });
            const addr = (p.address as number).toString(16).toUpperCase();
            return ok(`Wrote ${r.written} bytes → 0x${addr}`);
          }
    
          case "mgba_press_buttons": {
            const r = await mgba.call<{ queued: boolean; queue_size: number }>("press_buttons", {
              buttons:        p.buttons,
              frames:         p.frames         ?? 1,
              release_frames: p.release_frames ?? 1,
            });
            const keys = (p.buttons as string[]).join("+");
            return ok(
              `Queued press: ${keys} ` +
              `(hold ${p.frames ?? 1}f, release ${p.release_frames ?? 1}f). ` +
              `Queue size: ${r.queue_size}`,
            );
          }
    
          case "mgba_advance_frames": {
            const frame = await mgba.call<number>("advance_frames", { count: p.count ?? 1 });
            return ok(`Advanced ${p.count ?? 1} frame(s). Current frame: ${frame}`);
          }
    
          case "mgba_pause": {
            await mgba.call("pause");
            return ok("Emulation paused");
          }
    
          case "mgba_unpause": {
            await mgba.call("unpause");
            return ok("Emulation resumed");
          }
    
          case "mgba_reset": {
            await mgba.call("reset");
            return ok("ROM reset");
          }
    
          case "mgba_screenshot": {
            const path = await mgba.call<string>("screenshot", p.path ? { path: p.path } : {});
            return ok(`Screenshot saved: ${path}`);
          }
    
          case "mgba_save_state": {
            if (p.slot === undefined && p.path === undefined) {
              throw new Error("provide either `slot` (0-9) or `path`");
            }
            const r = await mgba.call<{ slot?: number; path?: string }>("save_state", {
              ...(p.slot !== undefined ? { slot: p.slot } : {}),
              ...(p.path !== undefined ? { path: p.path } : {}),
            });
            return ok(r.path ? `Saved state to ${r.path}` : `Saved state to slot ${r.slot}`);
          }
    
          case "mgba_load_state": {
            if (p.slot === undefined && p.path === undefined) {
              throw new Error("provide either `slot` (0-9) or `path`");
            }
            const r = await mgba.call<{ slot?: number; path?: string }>("load_state", {
              ...(p.slot !== undefined ? { slot: p.slot } : {}),
              ...(p.path !== undefined ? { path: p.path } : {}),
            });
            return ok(r.path ? `Loaded state from ${r.path}` : `Loaded state from slot ${r.slot}`);
          }
    
          default:
            throw new Error(`Unknown tool: ${name}`);
        }
      });
    }
  • MgbaClient.call() — the generic RPC helper used by the mgba_screenshot handler to invoke the 'screenshot' method on the mGBA bridge over a TCP socket.
      async call<T = unknown>(
        method: string,
        params?: Record<string, unknown>,
      ): Promise<T> {
        // Lazy (re)connect — bridge.lua reloads kill the socket, and the user
        // shouldn't have to restart the MCP host every time they edit the script.
        if (!this.socket || this.socket.destroyed) {
          try {
            await this.connect();
          } catch (err) {
            throw new Error(
              `Cannot reach mGBA bridge at ${this.host}:${this.port}. ` +
              `Make sure mGBA is running with bridge.lua loaded (Tools > Scripting). ` +
              `Underlying error: ${(err as Error).message}`,
            );
          }
        }
    
        return new Promise<T>((resolve, reject) => {
          const sock = this.socket;
          if (!sock) {
            reject(new Error("socket vanished after connect"));
            return;
          }
    
          const id = this.nextId++;
          this.pending.set(id, (resp) => {
            if (resp.error) {
              reject(new Error(`mGBA RPC error [${resp.error.code}]: ${resp.error.message}`));
            } else {
              resolve(resp.result as T);
            }
          });
    
          const msg = JSON.stringify({ id, method, params: params ?? {} }) + "\n";
          sock.write(msg, (err) => {
            if (err) {
              this.pending.delete(id);
              reject(err);
            }
          });
        });
      }
    }
Behavior3/5

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

No annotations provided, so the description must cover behaviors. It states saving to file and returning path, but does not mention potential side effects like pausing emulation or overwriting files. With no annotations, a score of 3 is adequate but not thorough.

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

Conciseness5/5

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

The description is a single, clear sentence with no wasted words. It front-loads the action and result efficiently.

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?

Given the tool's simplicity (one optional param, no output schema), the description is largely complete. It could mention prerequisites (e.g., emulation must be running) but is otherwise sufficient.

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

Parameters3/5

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

Schema description coverage is 100%, and the single parameter is described adequately. The description adds no extra meaning beyond the schema, so baseline 3 applies.

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 clearly states the tool's action: taking a screenshot, saving to file, and returning the path. It uses specific verbs and resources, and is distinct from sibling tools.

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

Usage Guidelines4/5

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

The description implies usage is for capturing the current display. While no explicit alternatives are given, sibling tools are all different operations, so confusion is minimal. However, it could explicitly state when to use this tool.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/dmang-dev/mcp-mgba'

If you have feedback or need assistance with the MCP directory API, please join our Discord server