ppsspp-dfx

Overview Schema Source Related Servers Discussions Admin

ppsspp_gpu_stats

Read-onlyIdempotent

Query GPU performance counters for an active PPSSPP session, retrieving fps, vblanks per second, and timing details to diagnose emulator performance issues.

Instructions

PURPOSE: Query GPU counters — fps, vblanks per second, timing info.

USAGE: session_id. The CPU must be RUNNING; paused, the MCP pre-probe returns CPU_STATE_ERROR instead of hanging — which doubles as the cheapest paused-CPU probe.

BEHAVIOR: READ-ONLY.

RETURNS: {fps, vblanks_per_second, info, timing, raw, text}.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`session_id`	Yes	Active session ID.

Output Schema

TableJSON Schema

Name	Required	Description
`fps`	Yes	Frames per second. None if PPSSPP didn't return it (e.g., no game running, or response shape differs).
`raw`	Yes	Raw `gpu.stats.get` response from PPSSPP.
`info`	Yes	GPU info dict (vendor / name / version, etc.).
`text`	Yes	Unified text representation: 'fps={FPS} vblanks={VBLANKS} info_keys={N} timing_keys={N}'.
`timing`	Yes	GPU timing dict (frame / block / vertex timing, etc.).
`vblanks_per_second`	Yes	VBlanks per second. None if PPSSPP didn't return it.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 16, 2026

TDQS

A4.5/5.0

Behavior5/5

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

Discloses a behavioral trait not covered by annotations: the MCP pre-probe returns CPU_STATE_ERROR when paused instead of hanging, which also doubles as a probe. The READ-ONLY statement is consistent with annotations and adds no contradiction, while the error behavior is genuinely additional context.

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 well-structured with PURPOSE, USAGE, BEHAVIOR, and RETURNS sections, each contributing value. Front-loaded with the purpose and behavior, it avoids filler while including a dense but useful note about the paused-CPU probe.

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?

For a single-parameter, read-only, idempotent tool with an output schema, the description covers purpose, usage condition, error behavior, and return fields. Nothing an agent needs to invoke it correctly is missing.

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?

The schema already fully documents the only parameter, session_id ('Active session ID.'), with 100% coverage. The description merely restates 'session_id' in the USAGE section without adding new semantic meaning, so the baseline of 3 is appropriate.

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?

Description states a specific verb ('Query GPU counters') and the exact metrics returned (fps, vblanks per second, timing info), leaving no ambiguity about the resource and action. It clearly differentiates from the sibling ppsspp_gpu_record by focusing on stats/query rather than recording.

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?

Provides explicit usage conditions: 'The CPU must be RUNNING' and explains the paused-CPU behavior (CPU_STATE_ERROR instead of hanging), including a clever secondary use as a cheapest paused-CPU probe. However, it does not name alternatives, so it falls short of full when-to-use vs. when-not-to-use guidance.

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