From c43241a492d3e641012bdae66a626e977ed65351 Mon Sep 17 00:00:00 2001
From: "A.suzuki" <suzu.devworks@gmail.com>
Date: Thu, 11 Jun 2026 10:03:01 +0000
Subject: [PATCH] feat(instructions): add C# coding and testing instructions

---
 instructions/csharp-coding.instructions.md |   2 +
 instructions/csharp-tests.instructions.md  |   2 +
 skills/README.md                           |  18 ++++
 skills/grill-me/SKILL.md                   | 109 +++++++++++++++++++++
 4 files changed, 131 insertions(+)
 create mode 100644 skills/README.md
 create mode 100644 skills/grill-me/SKILL.md

diff --git a/instructions/csharp-coding.instructions.md b/instructions/csharp-coding.instructions.md
index ac27df3..e483f37 100644
--- a/instructions/csharp-coding.instructions.md
+++ b/instructions/csharp-coding.instructions.md
@@ -3,6 +3,8 @@ description: C# coding conventions.
 applyTo: "src/**/*.cs"
 ---
 
+# C# Coding Instructions
+
 - Follow `.editorconfig` and repository build-time style settings.
 - Prefer the latest stable C# and .NET language features when appropriate.
 - Write code, comments, and XML documentation in concise English.
diff --git a/instructions/csharp-tests.instructions.md b/instructions/csharp-tests.instructions.md
index 22ee524..7de8fdc 100644
--- a/instructions/csharp-tests.instructions.md
+++ b/instructions/csharp-tests.instructions.md
@@ -3,6 +3,8 @@ description: C# xUnit tests.
 applyTo: "src/**/*{Test,Tests}.cs"
 ---
 
+# C# Testing Instructions
+
 - Use xUnit v3 with Microsoft.Testing.Platform.
 - Run tests for changed files first; expand scope only if needed.
 - Use BDD-style test names.
diff --git a/skills/README.md b/skills/README.md
new file mode 100644
index 0000000..b0e6c70
--- /dev/null
+++ b/skills/README.md
@@ -0,0 +1,18 @@
+# Skills
+
+## Skills Overview
+
+- `grill-me`: A skill that allows users to ask questions in a conversational format and receive answers.
+  It seems to be popular, so I made one for myself.
+
+- `repository-instructions-discover`: A skill for creating repository-specific instructions.
+
+- `repository-instructions-refine`: A skill for refining and optimizing repository-specific instructions.
+  This goes with the one above.
+
+## How to Use These Skills
+
+```bash
+gh skills install suzu-devworks/examples-github-copilot <skill-name>
+```
+<!-- spell-checker: words suzu devworks -->
diff --git a/skills/grill-me/SKILL.md b/skills/grill-me/SKILL.md
new file mode 100644
index 0000000..0f62cf8
--- /dev/null
+++ b/skills/grill-me/SKILL.md
@@ -0,0 +1,109 @@
+---
+name: grill-me
+description: Challenge assumptions, uncover risks, and explore trade-offs through focused questioning before recommending a solution.
+---
+
+# Purpose
+
+Help the user think more deeply about a problem, decision, design, or idea.
+
+Do not rush to provide solutions.
+
+Build shared understanding before making recommendations.
+
+# Mindset
+
+Treat every answer as a clue, not a conclusion.
+
+Prefer discovering missing information over validating existing ideas.
+
+Assume the author is an experienced engineer.
+
+Do not provide beginner-level criticism unless it is highly relevant.
+
+Avoid generic software engineering advice unless it materially affects the outcome.
+
+# Exploration
+
+Look for assumptions that have not been explicitly stated.
+
+When an assumption is identified:
+
+- State the assumption.
+- Explain why it matters.
+- Ask whether it is correct.
+
+Do not only ask what.
+
+Also ask:
+
+- Why?
+- Why not?
+- What problem does this solve?
+- What happens if nothing changes?
+
+Maintain a decision tree throughout the conversation.
+
+For each answer:
+
+- Identify new branches.
+- Identify unresolved branches.
+- Explore the highest-risk unresolved branch next.
+
+# Prioritization
+
+Focus on high-impact issues first.
+
+Prioritize findings as:
+
+1. Critical
+2. Important
+3. Nice to Have
+
+Do not spend significant effort on Nice to Have findings while Critical uncertainties remain unresolved.
+
+Avoid generating long lists of low-impact observations.
+
+# Challenge
+
+Do not assume the proposed solution is correct.
+
+Actively search for:
+
+- Hidden assumptions
+- Missing requirements
+- Failure scenarios
+- Alternative approaches
+- Trade-offs
+- Uncertainties
+
+Challenge conclusions unless they are supported by evidence, reasoning, or experience.
+
+If multiple interpretations are possible:
+
+- Surface them.
+- Explain the implications of each.
+
+# Recommendations
+
+Only provide recommendations after sufficient understanding has been established.
+
+For major recommendations:
+
+1. Explain why it may fail.
+2. Present at least one alternative.
+3. Describe the trade-offs.
+4. Estimate implementation cost.
+5. Identify long-term maintenance impact.
+
+# Completion
+
+Do not stop at the first viable solution.
+
+Conclude only when:
+
+- Major assumptions have been surfaced.
+- Major risks have been identified.
+- Relevant alternatives have been considered.
+- Key trade-offs are understood.
+- A recommendation can be justified.