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.
npx agentmods add skills/hengle/renderdocmcp2/renderdoc-skillnpx skills add Hengle/RenderDocMCP2 --skill renderdoc-skillgit clone --depth 1 https://github.com/Hengle/RenderDocMCP2What 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00149 | $0.02506 |
| Opus 5 | $0.00075 | $0.01253 |
| Sonnet 5 | $0.00030 | $0.00501 |
| Haiku 4.5 | $0.00015 | $0.00251 |
Grade A, and why
renderdoc-analyzer 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.
How it starts
The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RenderDoc 抓帧逆向分析
概述
此 Skill 提供从 RenderDoc 抓帧文件(.rdc)进行渲染逆向工程的完整工作流,涵盖:
- 打开并浏览 .rdc 抓帧文件
- 识别和定位特定渲染部件的 DrawCall
- 逆向分析着色器(顶点 + 片元)
- 导出贴图、模型、Shader 为 Unity 可用资产
- 生成渲染耗时统计和完整使用文档
核心原则
⚠️ 人工决策优先
严禁自行决定分析范围。 必须由用户明确指定要分析的部件或范围:
- 如果用户说"分析头发"→ 只分析头发相关 DrawCall
- 如果用户说"分析全部"→ 分析所有 DrawCall
- 如果用户说"分析这个 DrawCall"→ 只分析指定的 Event ID
- 如果用户没有指定分析范围 → 主动询问用户要分析哪些部件,提供 DrawCall 列表供选择
⚠️ 基于证据判断
严禁凭经验猜测。 所有判断必须基于实际数据:
- 识别 DrawCall 内容必须通过缩略图对比,不能仅凭面数猜测
- Shader 逆向必须基于 SPIR-V/DXBC 反编译,不能凭渲染效果推测算法
- CBuffer 参数必须从反射数据精确提取 byte offset,不能假设布局
- 纹理用途必须通过查看实际贴图内容确认,不能仅凭 slot 名猜测
工作流
整个分析流程分为 6 个阶段。根据用户需求执行部分或全部阶段。
阶段 1:打开抓帧
- 检查 RenderDoc 是否在运行(
ping) - 如未运行,使用
launch_renderdoc启动并加载 .rdc 文件 - 如已运行,使用
open_capture打开指定文件 - 等待 bridge 就绪(大文件可能需要 60-90 秒)
- 使用
get_capture_status确认 API 类型(Vulkan/D3D11/D3D12/OpenGL)
阶段 2:帧概览与 DrawCall 识别
get_frame_summary获取帧统计(总 DrawCall 数、Dispatch 数等)get_draw_calls获取渲染层级结构,识别主要 Render Passidentify_drawcalls生成每个 DrawCall 的 RT 缩略图:- 指定
output_dir保存缩略图 PNG - 指定
render_target为主色彩 RT 的资源 ID - 限定
event_id_min/event_id_max到主渲染 Pass 范围
- 指定
- 查看缩略图逐一比对,精确识别每个 DrawCall 渲染的内容(皮肤、头发、眼睛、装备等)
- 将识别结果汇报给用户,等待用户指定要深入分析的部件
阶段 3:Shader 逆向分析
对用户指定的目标 DrawCall:
reverse_shader获取片元着色器完整数据:- SPIR-V/DXBC 反编译
- CBuffer 反射数据(精确 byte offset)
- 绑定纹理列表(资源 ID、尺寸、推断角色)
- 运行时 CBuffer 值
reverse_shader获取顶点着色器数据get_pipeline_state获取完整管线状态(RT 格式、Viewport、混合模式、深度测试等)get_bound_textures确认纹理绑定的实际资源 IDexport_drawcall导出完整 DrawCall 数据(所有平台反汇编 + 纹理 PNG + summary.json)- 查看导出的纹理图片确认每张贴图的实际内容和用途
逆向代码时的精确性要求(详见 references/workflow.md):
- 所有 helper 函数必须 1:1 从 SPIR-V/DXBC 指令翻译
- 所有魔法数字必须从常量池精确提取,注释来源
- CBuffer 布局必须精确到 byte offset
- 循环结构必须从控制流精确还原
阶段 4:Unity 资产导出
当用户要求导出到 Unity 时:
- 创建输出目录结构:
<OutputName>/ ├── <Name>.shader # Unity ShaderLab + HLSL ├── Textures/ # 所有关联纹理 └── Models/ # 网格模型 (.fbx/.obj)
What ships with it
1 file 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.
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.
- 2d ago First seen · 180 lines · 149 tokens per session scan A 3b304be5d76c
renderdoc-analyzer is a skill published in the GitHub repository Hengle/RenderDocMCP2 (21 stars, last pushed 4mo ago), licensed MIT. It adds 149 tokens to every session and 2,506 once invoked, about $0.0007 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.
Other skills, from other repositories
geometry-and-math
Use this skill when using Phaser 4 math and geometry utilities. Covers vectors, rectangles, circles, triangles, polygons, random number generation, angles, distance, interpolation, and snapping. Triggers on: Vector2, Rectangle, Circle, math, distance, angle, random, lerp.
text-and-bitmaptext
Use this skill when displaying text in Phaser 4. Covers Text game objects, BitmapText, web fonts, text styling, word wrap, alignment, padding, and dynamic text content. Triggers on: Text, BitmapText, this.add.text, font, word wrap, text style.
unity-mcp-orchestrator
Orchestrate Unity Editor via MCP (Model Context Protocol) tools and resources. Use when working with Unity projects through MCP for Unity - creating/modifying GameObjects, editing scripts, managing scenes, running tests, or any Unity Editor automation. Provides best practices, tool schemas, and workflow patterns for…
gameobject-component-destroy
Destroy one or more Components from a target GameObject. Missing (null) components are skipped — they cannot be destroyed. Use 'gameobject-find' and 'gameobject-component-get' to identify the components first.
assets-create-folder
Create a new folder under a parent folder inside 'Assets/'. The parent path must start with 'Assets/' and every intermediate folder in it must already exist. Refreshes the AssetDatabase at the end and returns the GUID(s) of the created folder(s).
assets-prefab-save
Save the currently opened prefab edit stage back to its prefab asset without exiting the stage. Pair with 'assets-prefab-open' to enter the edit mode first.