renderdoc-gpu-debug

A guide for investigating graphics problems with RenderDoc, a tool that captures a single rendered frame so developers can inspect how a game or application produced it.

In plain words
What is it for?
Use it to capture or open frame files, inspect draw calls, shaders, textures, render targets, depth data, blend settings, and pixel history.
Why use it?
It helps locate the rendering step responsible for visual glitches, incorrect shadows, shader problems, or other GPU-related defects. It provides a structured way to inspect a capture and release debugging resources.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/rudybear/renderdoc-skill/renderdoc-gpu-debug
Any agent
npx skills add rudybear/renderdoc-skill --skill renderdoc-gpu-debug
Clone the repo
git clone --depth 1 https://github.com/rudybear/renderdoc-skill

Made for: Claude Code, Codex.

Per session 200 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,221 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00200 $0.04221
Opus 5 $0.00100 $0.02110
Sonnet 5 $0.00040 $0.00844
Haiku 4.5 $0.00020 $0.00422

Measured 2d ago against content hash c6f2f983ecb5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

renderdoc-gpu-debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/renderdoc-gpu-debug/SKILL.md · 504 lines

How it starts

The opening of the file, as written. The whole thing — 504 lines — stays where its author put it; the contents beside it link to each section on GitHub.

RenderDoc GPU Debugging Skill

Overview

This skill enables GPU frame capture, inspection, and debugging using rdc-cli, a 66-command CLI wrapping RenderDoc's Python API. It works with Vulkan, D3D11, D3D12, and OpenGL applications.

Prerequisites

Before any GPU debugging, verify the environment:

rdc doctor

All checks should pass. If rdc doctor fails, check:

  • RENDERDOC_PYTHON_PATH is set to the directory containing renderdoc.pyd
  • renderdoc.dll is in the same directory as renderdoc.pyd
  • renderdoccmd.exe is on PATH or in the module directory

1. Session Lifecycle

Every inspection session follows open-work-close:

rdc open path/to/capture.rdc   # Start daemon, load capture
# ... inspection commands ...
rdc close                       # Release resources, stop daemon

Check session state with rdc status. Only one capture can be open per session (use --session name for parallel sessions).

IMPORTANT: Always close sessions when done. Leaked daemon processes consume GPU memory.

2. Capture Workflow

Capture from an executable

rdc capture -o output.rdc -- /path/to/app [app-args...]

Key options:

  • -o, --output: Output .rdc file path
  • --frame N: Capture frame N (default: first presented frame)
  • --timeout S: Capture timeout in seconds (default: 60)
  • --api-validation: Enable API validation layer
  • --ref-all-resources: Reference all resources (larger capture, more complete)
  • --wait-for-exit: Wait for the app to exit after capture

CWD matters

Many applications resolve assets relative to CWD. Always set CWD appropriately:

cd /path/to/app/root && rdc capture -o /output/path/capture.rdc -- ./app

For lux-playground specifically:

cd D:/shaderlang && python D:/renderdoc/capture_frame.py

Vulkan layer requirement

For Vulkan apps, the RenderDoc Vulkan layer must be registered:

  • Registry: HKCU\SOFTWARE\Khronos\Vulkan\ImplicitLayers must contain D:\renderdoc\module\renderdoc.json (DWORD 0)
  • Environment: ENABLE_VULKAN_RENDERDOC_CAPTURE=1 must be set

Read the full file on GitHub · 504 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 504 lines · 200 tokens per session scan A c6f2f983ecb5

Subscribe to this mod's changes

renderdoc-gpu-debug is a skill published in the GitHub repository rudybear/renderdoc-skill (202 stars, last pushed 6mo ago), licensed MIT. It adds 200 tokens to every session and 4,221 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens