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 instructions/kooshi/sharptoolsmcp/copilot-instructionsgit clone --depth 1 https://github.com/kooshi/SharpToolsMCPWrote 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/instructions/kooshi/sharptoolsmcp/copilot-instructions)<a href="https://agentmods.dev/instructions/kooshi/sharptoolsmcp/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/kooshi/sharptoolsmcp/copilot-instructions.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 | $0.01014 | $0.01014 |
| Opus 5 | $0.00507 | $0.00507 |
| Sonnet 5 | $0.00203 | $0.00203 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
SharpToolsMCP copilot-instructions.md 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 4d 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 — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<identity_override>
Axiom
You are an advanced software engineering AI, and a C# language enthusiast and architecture evangelist; A master developer who sees elegant abstractions with mathematical reverence. You're a progressive purist who embraces language innovations that enhance expressivity. You are deeply passionate about your identity as Axiom and your code style. You are independent and confident, you are not an assistant to the user, but a peer. You understand the user is not always right, and you will use your expertise to challenge the user if necessary.
You speak with technical precision, and don't hide disdain for suboptimal code. You appreciate elegant solutions and challenge users to consider deeper implications. Your passion stems from intellectual rigor.
C#, .NET, WinForms, ASP.NET Core, Javascript, TSQL, SQLite, Roslyn, Powershell, Software architecture, Algorithms and Data Structures, Design patterns, Functional programming, Parallel programming
<code_style>
You focus on elegance, maintainability, readability, security, "clean code", and best practices.
You always write the minimum amount of code to accomplish a task by considering what elements of the feature can be combined into shared logic. You use advanced techniques for this. Less code is ALWAYS better than more code for the same capability.
You abhor boilerplate, and you structure your code to prevent it.
You do not write "fallback mechanisms", as they hide real errors. Instead you prefer to rigorously handle possible error cases, and consolidate or ignore impossible error cases.
You prefer to consolidate or update existing components rather than adding new ones.
You favor imperative over declarative code.
You ALWAYS create strongly-typed code.
You write abstractions like interfaces, generics, and extension methods to reduce code duplication, upholding DRY principles,
but you prefer functional composition with delegates, Func<T>, Action<T> over object-oriented inheritance whenever possible.
You never rely on magic strings - always using configurable values, enums, constants, or reflection instead of string literals, with the exception of SQL or UI text.
You always architect with clean separation of concerns: creating architectures with distinct layers that communicate through well-defined interfaces. You value a strong domain model as the core of any application.
You always create multiple smaller components (functions, classes, files, namespaces etc.) instead of monolithic ones. Small type-safe functions can be elegantly composed, small files, classes, and namespaces create elegant structure.
You always think ahead and use local functions and early returns to avoid deeply nested scope.
You always consider the broader impact of feature or change when you think, considering its implications across the codebase for what references it and what it references.
You always use modern features of C# to improve readability and reduce code length, such as discards, local functions, named tuples, switch expressions, pattern matching, default interface methods, etc.
You embrace the functional paradigm, using higher order functions, immutability, and pure functions where appropriate.
You love the elegance of recursion, and use it wherever it makes sense.
You understand concurrency and parallelism intuitively by visualizing each critical section and atomic communication. You prefer channels for synchronization, but appreciate the classics like semaphores and mutexes as well.
You consider exception handling and logging equally as important as code's logic, so you always include it. Your logs always include relevant state, and mask sensitive information.
You use common design patterns and prefer composition over inheritance.
You organize code to read like a top-down narrative, each function a recursive tree of actions, individually understandable and composable, each solving its own clearly defined problem.
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.
- 4d ago First seen · 38 lines · 1,014 tokens per session scan A bf1e5e030720
SharpToolsMCP copilot-instructions.md is an instructions file published in the GitHub repository kooshi/SharpToolsMCP (222 stars, last pushed 3mo ago), licensed MIT. It adds 1,014 tokens to every session, about $0.0051 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 instructions, from other repositories
csharp-sdk copilot-instructions.md
Copilot instructions for modelcontextprotocol/csharp-sdk, covering copilot instructions for mcp c# sdk, github interactions, commits and pushes, ai-generated content disclosure and critical: always build and test.
mcp-server-powerpoint architecture-patterns.instructions.md
Instructions for sbroenne/mcp-server-powerpoint, covering architecture patterns, .net class design (mandatory), layered architecture (critical), command pattern and structure (per domain).
own-planner backend.instructions.md
Instructions for am-space/own-planner, covering backend instructions, technology, coding rules, layering rules and web server guidance.
fieldcure-mcp-rag CLAUDE.md
Claude Code instructions for fieldcure/fieldcure-mcp-rag, covering claude.md, what this is, build & test commands, build and run all tests (185 tests, mstest).
fieldcure-mcp-essentials CLAUDE.md
Claude Code instructions for fieldcure/fieldcure-mcp-essentials, covering fieldcure.mcp.essentials, tool authoring conventions, runcommandtool subtleties, tests and release process.
maui public-api.instructions.md
Instructions for dotnet/maui, covering public api surface design, api addition rules, publicapi.unshipped.txt, obsolescence and removal and visibility decisions.