Skip to main content
Glama

save_project

Save the open FMOD Studio project to preserve changes before building audio banks.

Instructions

Save the open FMOD Studio project. Must be called before build_banks.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • Core handler that executes 'save_project'. Sends JavaScript `studio.project.save()` to FMOD Studio via TCP client.
    async def save_project(client: StudioClient) -> dict[str, Any]:
        """Save the currently open FMOD Studio project."""
        js = """
            studio.project.save();
            return { ok: true, saved: true };
        """
        return await client.eval(js, timeout=120.0)
  • Registers 'save_project' as an MCP tool on the FastMCP server using the @mcp.tool() decorator.
    @mcp.tool()
    async def save_project() -> dict[str, Any]:
        """Save the open FMOD Studio project. Must be called before build_banks."""
        return await project.save_project(_studio())
  • Registration test confirms 'save_project' is listed in the expected tools set.
        "save_project",
        "build_banks",
        # escape
        "run_js",
    }
    
    
    def test_all_tools_registered():
        tools = set(mcp._tool_manager._tools.keys())
        missing = EXPECTED_TOOLS - tools
        extra = tools - EXPECTED_TOOLS
        assert not missing, f"Missing tools: {missing}"
        assert not extra, f"Unexpected extra tools: {extra}"
  • StudioClient.eval() is the helper used by save_project to execute JavaScript in FMOD Studio. It wraps JS in an IIFE with sentinel-based response detection over TCP.
    async def eval(self, js: str, timeout: float = 30.0) -> Any:
        """Evaluate a JavaScript snippet in Studio and return the parsed value.
    
        The caller is responsible for making sure ``js`` uses ``return`` to
        surface its value (the snippet is wrapped in an IIFE).
        """
        req_id = uuid.uuid4().hex
        # Substitute the request id into the template BEFORE injecting user JS,
        # so a snippet that happens to contain "__ID__" or "__JS__" cannot
        # corrupt the surrounding wrapper.
        wrapped = _WRAPPER_TEMPLATE.replace("__ID__", req_id).replace("__JS__", js)
        # Studio's terminal evaluates per-line: strip `//` comments and collapse
        # internal newlines so the whole command arrives as one line.
        wrapped = _flatten_js(wrapped)
        async with self._lock:
            # connect() inside the lock so concurrent callers don't open
            # multiple sockets or fight over the same StreamReader.
            await self.connect()
            self._log_command(req_id, js)
            assert self._writer is not None
            self._writer.write(wrapped.encode("utf-8"))
            await self._writer.drain()
            return await self._await_response(req_id, timeout)
    
    async def _await_response(self, req_id: str, timeout: float) -> Any:
        assert self._reader is not None
        loop = asyncio.get_event_loop()
        deadline = loop.time() + timeout
        while True:
            found = self._consume_response(req_id)
            if found is not None:
                kind, payload = found
                if kind == "OK":
                    return payload
                message = payload.get("message", "FMOD Studio error") if isinstance(payload, dict) else str(payload)
                stack = payload.get("stack", "") if isinstance(payload, dict) else ""
                raise StudioError(f"{message}\n{stack}".rstrip())
            remaining = deadline - loop.time()
            if remaining <= 0:
                raise TimeoutError(
                    f"FMOD Studio did not respond to request {req_id} within {timeout}s"
                )
            try:
                chunk = await asyncio.wait_for(self._reader.read(4096), timeout=remaining)
            except asyncio.TimeoutError as exc:
                raise TimeoutError(
                    f"FMOD Studio did not respond to request {req_id} within {timeout}s"
                ) from exc
            if not chunk:
                raise ConnectionError("FMOD Studio closed the TCP connection")
            self._buffer += chunk.decode("utf-8", errors="replace")
    
    def _consume_response(self, req_id: str) -> tuple[str, Any] | None:
        for kind, sentinel in (("OK", _SENTINEL_OK), ("ERR", _SENTINEL_ERR)):
            tag = f"{sentinel}:{req_id}:"
            idx = self._buffer.find(tag)
            if idx == -1:
                continue
            end = self._buffer.find("\n", idx + len(tag))
            if end == -1:
                return None
            payload_str = self._buffer[idx + len(tag) : end]
            self._buffer = self._buffer[end + 1 :]
            try:
                return kind, json.loads(payload_str)
            except json.JSONDecodeError as exc:
                raise StudioError(f"Malformed JSON from Studio: {payload_str!r}") from exc
        return None
    
    def _log_command(self, req_id: str, js: str) -> None:
        try:
            with self._log_path.open("a", encoding="utf-8") as fh:
                fh.write(f"# {req_id}\n{js.rstrip()}\n\n")
        except OSError as exc:
            logger.warning("Could not write to command log %s: %s", self._log_path, exc)
  • Unit test for save_project, verifying it calls 'studio.project.save' and returns saved=True.
    async def test_save_project(client: StudioClient, mock_studio: MockStudio):
        mock_studio.responder = responder_sequence([("OK", {"ok": True, "saved": True})])
        result = await project.save_project(client)
        assert result["saved"] is True
        assert "studio.project.save" in _last_sent_js(mock_studio)
Behavior3/5

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

No annotations provided, so the description carries full burden. It states the save action but does not disclose whether it overwrites, is idempotent, or has any side effects. Minimal but adequate for a simple save.

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?

Two short sentences with no fluff. The purpose is front-loaded, and the usage condition follows immediately.

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

Completeness5/5

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

Despite no annotations and no parameters, the description fully covers the save operation's purpose and prerequisite. An output schema exists, so return values are handled separately.

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?

No parameters exist in the schema, and schema coverage is 100%. The description adds no param details, but baseline 4 is appropriate as there is nothing to add.

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 verb 'Save' and the resource 'the open FMOD Studio project', distinguishing it from sibling tools like build_banks or list_banks.

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?

Explicitly states the prerequisite: 'Must be called before build_banks', providing clear guidance on when to use this tool relative to others.

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/jmperez127/fmod-mcp'

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