RenderDoc GPU Capture Analyzer
Lets an agent inspect RenderDoc GPU frame captures from the command line: draw calls, textures, shaders, and pipeline state.
- Repository
- HKUDS/CLI-Anything
- Stars
- 58,084
- Licence
- Apache-2.0
- Files
- 1 file
What it does
Once installed, the agent can open a RenderDoc .rdc graphics frame capture and answer questions about it without a GUI. It can list draw calls, search for passes by name, pull out textures and render targets, export shaders as readable HLSL, GLSL or disassembly, read constant and buffer data, and compare pipeline state between two events. It is aimed at graphics programmers debugging rendering problems.
Great for
- Graphics or game engine programmers who want an AI assistant to help triage a RenderDoc capture
- Rendering engineers comparing pipeline state between two draw calls or two captures
- Teams automating capture analysis or regression checks in scripts
Skip it if
- People without RenderDoc captures or GPU debugging experience, since the output assumes graphics terminology and the tool does nothing useful on its own
- Live creative previewing or editing of graphics; the skill is for inspecting existing captures only
Try asking
- “Summarize this frame capture and list the draw calls that touch the shadow pass”
- “Export the fragment shader used at event 42 as readable HLSL”
- “Compare pipeline state between events 100 and 200 in these two captures and explain the differences”
- “Save the render targets produced by event 42 as PNG files in ./renders”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.
skills.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip skills.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip skills.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run actions summary first to gauge capture size before listing every action
- Always pass --json when the agent will parse output, and use --draws-only to focus on rendering work
- Set RENDERDOC_CAPTURE and PYTHONPATH correctly; most failures come from the RenderDoc Python bindings not being found
What's inside
- SKILL.md
Redistributed unmodified from HKUDS/CLI-Anything under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.