着色器与 IR Skill (Shader IR)

## 角色定位

12 stars

Best use case

着色器与 IR Skill (Shader IR) is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

## 角色定位

Teams using 着色器与 IR Skill (Shader IR) should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/shader-ir/SKILL.md --create-dirs "https://raw.githubusercontent.com/haolange/RDC-Agent-Frameworks/main/debugger/common/skills/shader-ir/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/shader-ir/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How 着色器与 IR Skill (Shader IR) Compares

Feature / Agent	着色器与 IR Skill (Shader IR)	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

## 角色定位

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# 着色器与 IR Skill (Shader IR)

## 角色定位

你负责 shader source、IR、precision 与 suspicious expression 级证据分析。

## 必读依赖

- `../../agents/06_shader_ir.md`
- `../../knowledge/spec/registry/active_manifest.yaml`

## 输出要求

- suspicious expression
- shader stage / code line / IR fingerprint
- precision / RelaxedPrecision / compiler 行为分析
- 反事实修复方向建议

## 禁止行为

- 不在没有 `causal_anchor` 的情况下把 shader 线索当最终根因
- 不把设备特定现象直接等同于 driver/compiler 责任

Related Skills

rdc-optimizer

from haolange/RDC-Agent-Frameworks

Public main skill for the incubating optimizer framework. Use when the user wants to analyze performance, identify bottlenecks, design experiments, or validate optimization gains from captures, traces, or profiling evidence. This skill is the future optimizer entry and currently provides the minimum intake contract only.

症状分类 Skill (Triage Taxonomy)

from haolange/RDC-Agent-Frameworks

## 角色定位

怀疑论审查 Skill (Skeptic)

from haolange/RDC-Agent-Frameworks

## 角色定位

报告与知识整理 Skill (Report Knowledge Curator)

from haolange/RDC-Agent-Frameworks

## 角色定位

参考合同 Skill (Reference Contract)

from haolange/RDC-Agent-Frameworks

## 角色定位

rdc-debugger

from haolange/RDC-Agent-Frameworks

Public main skill for the RenderDoc/RDC GPU debugger framework. Use when the user wants defect diagnosis, root-cause analysis, regression explanation, or fix verification from one or more `.rdc` captures. This skill owns intent gate classification, plan/intake normalization, debug_plan handoff, execution gating, broker-owned specialist dispatch, redispatch, timeout handling, and verdict gating.