rdc-analyst

Name: rdc-analyst
Author: haolange

15stars5forks

How to install rdc-analyst?

haolange/rdc-agentcaps/rdc-analyst

$npx -y skills add haolange/RDC-AgentCaps --skill rdc-analyst

Installs into the current project.

Summary

Analyze GPU captures and frame data from RenderDoc (.rdc files), PIX, or similar frame debuggers. Reconstructs render pass graphs, resource dependency chains, material/shader bindings, and pipeline state for a captured frame. Use when the user wants to understand how a frame is r

Files · 1

View on GitHub

SKILL.md

1	# RDC Analyst
2
3	Analyze GPU frame captures to reconstruct render pass graphs, resource flows, and engine structure. This skill handles analysis and knowledge-building requests; for defect debugging, use the debugger framework instead.
4
5	## Workflow
6
7	### Step 1: Identify the Input
8
9	Determine what capture data is available:
10
11	\| Input Type \| Examples \| Action \|
12	\|------------\|----------\|--------\|
13	\| RenderDoc capture \| `.rdc` file, texture/buffer viewer output \| Proceed with pass-level analysis \|
14	\| Frame debugger log \| PIX capture, Xcode GPU trace, NVIDIA Nsight export \| Proceed — normalize to pass/draw-call model \|
15	\| Engine replay data \| Unreal Insights, Unity Frame Debugger dump \| Proceed — map to render pass abstraction \|
16	\| Text description only \| "We have 3 passes: shadow, gbuffer, lighting" \| Proceed with user-provided structure \|
17	\| No capture available \| User has no data yet \| Help user plan what to capture and with which tool \|
18
19	### Step 2: Clarify the Analysis Goal
20
21	Ask the user to confirm what they need. If unclear, ask explicitly:
22
23	1. What is the capture or input? (file type, engine, API — D3D12, Vulkan, Metal, etc.)
24	2. What analysis product do you need?
25	- Pass graph (render pass sequence with dependencies)
26	- Resource dependency chain (which textures/buffers flow between passes)
27	- Material/shader structure map (bindings, permutations, parameter sources)
28	- Pipeline state summary (blend modes, rasterizer config per draw)
29	- Knowledge entry (reusable reference doc for this rendering technique)
30	3. What scope? Full frame, a specific pass range, or a single draw call?
31
32	### Step 3: Perform the Analysis
33
34	Follow the appropriate path based on the requested product:
35
36	Pass graph reconstruction:
37	1. List all render passes in execution order
38	2. For each pass, record: name/label, render targets, input resources, draw count
39	3. Identify dependencies (pass B reads a resource written by pass A)
40	4. Output as a structured table or diagram description
41
42	Resource dependency chain:
43	1. Enumerate key resources (render targets, depth buffers, UAVs)
44	2. Track producer and consumer passes for each resource
45	3. Flag resources that are written but never read (potential waste)
46
47	Material/shader analysis:
48	1. Group draw calls by pipeline state object or shader combination
49	2. Identify material parameter sources (constant buffers, push constants)
50	3. Note shader permutation patterns and binding slot conventions
51
52	Knowledge entry:
53	1. Synthesize findings into a reusable reference document
54	2. Include: technique name, pass sequence, resource flow, shader roles, key parameters
55	3. Format for storage in `analyzer/common/` as a knowledge artifact
56
57	### Step 4: Validate and Deliver
58
59	Before delivering results:
60	- [ ] Every claim references a specific pass, draw call, or resource by name/index
61	- [ ] Dependencies are directional (producer -> consumer) and acyclic where expected
62	- [ ] Output format matches what the user requested
63	- [ ] Scope matches what was agreed in Step 2
64
65	## Routing Table
66
67	\| User Says \| Route To \|
68	\|-----------\|----------\|
69	\| "Why is this triangle wrong" / "rendering bug" / "visual glitch" \| Debugger — this is a defect, not analysis \|
70	\| "Map out the render passes" / "show me the pass graph" \| This skill — pass graph reconstruction \|
71	\| "What resources does the shadow pass use" \| This skill — resource dependency chain \|
72	\| "How does this engine set up materials" \| This skill — material/shader analysis \|
73	\| "Document this rendering technique" \| This skill — knowledge entry \|
74
75	## Example
76
77	User: "I have a RenderDoc capture of our deferred renderer. Can you map out the pass structure and resource flow?"
78
79	Intake response:
80	> To map the pass structure, I need to confirm:
81	> 1. API: Which graphics API — D3D12, Vulkan, or Metal?
82	> 2. Scope: Full frame, or a specific section (e.g., just the lighting passes)?
83	> 3. Output: Pass graph with resource dependencies, or also material/shader breakdown?
84	>
85	> Once confirmed, I will reconstruct the pass sequence, identify render targets and input resources per pass, and produce a dependency table showing resource flow across passes.
86
87	After intake — sample output structure:
88
89	\| Pass \| Render Targets \| Input Resources \| Draw Calls \|
90	\|------\|---------------\|-----------------\|------------\|
91	\| Shadow Depth \| ShadowMap (D32) \| — \| 142 \|
92	\| GBu

People who liked this also viewed

6 picks

Type

rdc-analyst

15stars5forks

View on GitHub

How to install rdc-analyst?

haolange/rdc-agentcaps/rdc-analyst

$npx -y skills add haolange/RDC-AgentCaps --skill rdc-analyst

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# RDC Analyst
2
3	Analyze GPU frame captures to reconstruct render pass graphs, resource flows, and engine structure. This skill handles analysis and knowledge-building requests; for defect debugging, use the debugger framework instead.
4
5	## Workflow
6
7	### Step 1: Identify the Input
8
9	Determine what capture data is available:
10
11	\| Input Type \| Examples \| Action \|
12	\|------------\|----------\|--------\|
13	\| RenderDoc capture \| `.rdc` file, texture/buffer viewer output \| Proceed with pass-level analysis \|
14	\| Frame debugger log \| PIX capture, Xcode GPU trace, NVIDIA Nsight export \| Proceed — normalize to pass/draw-call model \|
15	\| Engine replay data \| Unreal Insights, Unity Frame Debugger dump \| Proceed — map to render pass abstraction \|
16	\| Text description only \| "We have 3 passes: shadow, gbuffer, lighting" \| Proceed with user-provided structure \|
17	\| No capture available \| User has no data yet \| Help user plan what to capture and with which tool \|
18
19	### Step 2: Clarify the Analysis Goal
20
21	Ask the user to confirm what they need. If unclear, ask explicitly:
22
23	1. What is the capture or input? (file type, engine, API — D3D12, Vulkan, Metal, etc.)
24	2. What analysis product do you need?
25	- Pass graph (render pass sequence with dependencies)
26	- Resource dependency chain (which textures/buffers flow between passes)
27	- Material/shader structure map (bindings, permutations, parameter sources)
28	- Pipeline state summary (blend modes, rasterizer config per draw)
29	- Knowledge entry (reusable reference doc for this rendering technique)
30	3. What scope? Full frame, a specific pass range, or a single draw call?
31
32	### Step 3: Perform the Analysis
33
34	Follow the appropriate path based on the requested product:
35
36	Pass graph reconstruction:
37	1. List all render passes in execution order
38	2. For each pass, record: name/label, render targets, input resources, draw count
39	3. Identify dependencies (pass B reads a resource written by pass A)
40	4. Output as a structured table or diagram description
41
42	Resource dependency chain:
43	1. Enumerate key resources (render targets, depth buffers, UAVs)
44	2. Track producer and consumer passes for each resource
45	3. Flag resources that are written but never read (potential waste)
46
47	Material/shader analysis:
48	1. Group draw calls by pipeline state object or shader combination
49	2. Identify material parameter sources (constant buffers, push constants)
50	3. Note shader permutation patterns and binding slot conventions
51
52	Knowledge entry:
53	1. Synthesize findings into a reusable reference document
54	2. Include: technique name, pass sequence, resource flow, shader roles, key parameters
55	3. Format for storage in `analyzer/common/` as a knowledge artifact
56
57	### Step 4: Validate and Deliver
58
59	Before delivering results:
60	- [ ] Every claim references a specific pass, draw call, or resource by name/index
61	- [ ] Dependencies are directional (producer -> consumer) and acyclic where expected
62	- [ ] Output format matches what the user requested
63	- [ ] Scope matches what was agreed in Step 2
64
65	## Routing Table
66
67	\| User Says \| Route To \|
68	\|-----------\|----------\|
69	\| "Why is this triangle wrong" / "rendering bug" / "visual glitch" \| Debugger — this is a defect, not analysis \|
70	\| "Map out the render passes" / "show me the pass graph" \| This skill — pass graph reconstruction \|
71	\| "What resources does the shadow pass use" \| This skill — resource dependency chain \|
72	\| "How does this engine set up materials" \| This skill — material/shader analysis \|
73	\| "Document this rendering technique" \| This skill — knowledge entry \|
74
75	## Example
76
77	User: "I have a RenderDoc capture of our deferred renderer. Can you map out the pass structure and resource flow?"
78
79	Intake response:
80	> To map the pass structure, I need to confirm:
81	> 1. API: Which graphics API — D3D12, Vulkan, or Metal?
82	> 2. Scope: Full frame, or a specific section (e.g., just the lighting passes)?
83	> 3. Output: Pass graph with resource dependencies, or also material/shader breakdown?
84	>
85	> Once confirmed, I will reconstruct the pass sequence, identify render targets and input resources per pass, and produce a dependency table showing resource flow across passes.
86
87	After intake — sample output structure:
88
89	\| Pass \| Render Targets \| Input Resources \| Draw Calls \|
90	\|------\|---------------\|-----------------\|------------\|
91	\| Shadow Depth \| ShadowMap (D32) \| — \| 142 \|
92	\| GBu

People who liked this also viewed

6 picks

Type