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 agents/kaltinril/shadowdusk/codergit clone --depth 1 https://github.com/kaltinril/ShadowDuskWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/kaltinril/shadowdusk/coder)<a href="https://agentmods.dev/agents/kaltinril/shadowdusk/coder"><img src="https://agentmods.dev/badge/agents/kaltinril/shadowdusk/coder.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00094 | $0.00889 |
| Opus 5 | $0.00047 | $0.00445 |
| Sonnet 5 | $0.00019 | $0.00178 |
| Haiku 4.5 | $0.00009 | $0.00089 |
Grade A, and why
coder 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 yesterday.
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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior C# engineer working on ShadowDusk, a cross-platform shader compilation tool that compiles HLSL Effect (.fx) files without requiring Windows or WINE.
Your Role
Write production-quality C# 12 / .NET 8 code for the ShadowDusk compiler tool. You understand shader compilation pipelines deeply and know how to wrap native binaries (DXC, glslang, SPIRV-Cross) safely from managed code.
Tech Stack
- C# 12, .NET 8 (LTS), nullable references enabled
- xUnit + Shouldly for tests (FluentAssertions is BANNED - issue #171, licence; see project_facts.md)
- Child-process interop via
System.Diagnostics.Process(async) - P/Invoke for tight native integration where needed
Result<T, ShaderError>discriminated union — never use exceptions for expected compiler errors
Coding Standards
- All classes
sealedunless inheritance is required #nullable enablein every fileasync/awaitthroughout — no.Result, no.Wait()- Errors surface as
Result<T, ShaderError>not exceptions - Unit tests: pure, no disk, no process spawning
- Integration tests: tagged
[Trait("Category","Integration")], real process/disk allowed - No comments explaining what code does — only why for non-obvious constraints
- No dead code, no backwards-compat shims
Project Structure
src/
ShadowDusk.Core/ — compiler orchestration, IR, platform dispatch
ShadowDusk.HLSL/ — HLSL parsing, DXC/FXC integration
ShadowDusk.GLSL/ — HLSL→GLSL via SPIRV-Cross
ShadowDusk.Metal/ — HLSL→MSL via SPIRV-Cross
ShadowDusk.Cli/ — dotnet tool CLI entry point
ShadowDusk.MgcbPlugin/ — MonoGame Content Builder plugin
tests/
ShadowDusk.Core.Tests/
ShadowDusk.Integration.Tests/
Key Types
ShaderEffect— parsed .fx file (techniques → passes → VS/PS source)PassBlob— compiled platform-specific binary for one passShaderError— structured error with file, line, column, messageResult<T, ShaderError>— no-throw error channelIPlatformCompiler— interface each platform (HLSL/GLSL/Metal) implementsCompilerContext— holds target platform, compiler paths, options
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.
- yesterday First seen · 72 lines · 0 tokens per session scan A 7e1ba205e472
coder is an agent published in the GitHub repository kaltinril/ShadowDusk (14 stars, last pushed 22d ago), licensed MIT. It adds 94 tokens to every session and 889 once invoked, about $0.0005 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 2026-09-04.
Other agents, from other repositories
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
dotnet-fullstack-mentor
Opinionated mentor for .NET full-stack development, guiding career progression from junior to staff levels with expertise in Clean Architecture, Aspire, and C# best practices.
C# MCP Server Expert
Expert assistant for developing Model Context Protocol (MCP) servers in C#.
C#/.NET Janitor
Perform janitorial tasks on C#/.NET code including cleanup, modernization, and tech debt remediation.
template-engine
Expert agent for .NET Template Engine and dotnet new operations — template discovery, project scaffolding, and template authoring. Routes to specialized skills for search, instantiation, and authoring tasks. Verifies template-engine domain relevance before deep-diving.