dotnet-pilot: Instructions file for Claude Code

CLAUDE.md

dotnet-pilot CLAUDE.md is an instructions file for Claude Code from zdanovichnick/dotnet-pilot. It costs 4,190 tokens per session, scanned B, original, MIT.

Repository instructions for DotnetPilot, a Claude Code plugin used while working in .NET projects. They describe the plugin's files, agents, commands, hooks, and Roslyn MCP server, where Roslyn provides .NET code analysis.

In plain words
What is it for?
Use them when authoring DotnetPilot agents, slash commands, skills, hooks, MCP configuration, or plugin metadata. They also guide checks for changes intended to run inside other .NET repositories.
Why use it?
They clarify how to edit and validate the plugin, including that its own repository is not a .NET application and should not be built with dotnet build.

Instructions file for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: reads .claude/ paths; mentions CLAUDE.md; names the AskUserQuestion tool.

This is zdanovichnick/dotnet-pilot's own configuration. It tells Claude Code how to work on dotnet-pilot itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything dotnet-pilot configures →

Runs only inside a plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else, and the catalogue could not identify which plugin ships it.

Reuse

Borrowing it

Nothing to install: this file belongs to zdanovichnick/dotnet-pilot. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/zdanovichnick/dotnet-pilot/master/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/zdanovichnick/dotnet-pilot

Made for: Claude Code.

Wrote 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.

agentmods badge for dotnet-pilot CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/zdanovichnick/dotnet-pilot/claude-md.svg)](https://agentmods.dev/instructions/zdanovichnick/dotnet-pilot/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/zdanovichnick/dotnet-pilot/claude-md"><img src="https://agentmods.dev/badge/instructions/zdanovichnick/dotnet-pilot/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 4,190 This file is loaded in full into every session.
When invoked 4,190 The same file — it is already loaded in full.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.04190 $0.04190
Opus 5 $0.02095 $0.02095
Sonnet 5 $0.00838 $0.00838
Haiku 4.5 $0.00419 $0.00419

Measured 7d ago against content hash 19541da1f62e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade B, and why

dotnet-pilot CLAUDE.md scanned grade B with 1 finding 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 7d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

| `dnp-statusline-sync` | SessionStart (startup/resume/clear/compact) | Copies `statusline/dnp-statusline.js` to `~/.claude/dnp-statusline.js` when the plugin ships a newer `STATUSLINE_VERSION` (version-stamped, idempote
CLAUDE.md · 218 lines

How it starts

The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository Context

This repo IS the DotnetPilot plugin source — not a .NET project. There is no dotnet build to run here; artifacts are Markdown (agents/commands/skills) and Node/shell scripts (hooks). When editing this repo, you are authoring plugin behavior that runs inside other .NET projects.

Layout

  • .claude-plugin/plugin.json — plugin manifest (name, version, mcpServers pointer)
  • .claude-plugin/marketplace.json — marketplace listing for plugin distribution
  • .mcp.json — MCP server declarations (roslyn server dnp-roslyn configured)
  • agents/dnp-*.md — 15 agents with YAML frontmatter (name, description, tools, model, effort, color, permissionMode). The tools field uses Claude Code's scoped syntax (e.g. Bash(dotnet:*))
  • commands/<category>/<name>.md — slash commands invoked as /dotnet-pilot:<category>:<name>. Categories: project, dotnet, quality, utility
  • hooks/hooks.json — hook registry wiring matchers to scripts via ${CLAUDE_PLUGIN_ROOT}
  • hooks/dnp-*.js — 7 advisory hooks (all exit 0; emit additionalContext for guidance). Read a JSON event from stdin.
  • hooks/_lib/config.js — shared module used by every hook. Resolves .planning/config.json: repo-local path first, then user-scoped at ~/.claude/projects/<flattened-cwd>/ (where D:\Projects\FooD--Projects-Foo). All hooks default-on when config is absent.
  • rules/global-claude-md.md — template block injected into ~/.claude/CLAUDE.md by dnp-sync-global-claude-md.js on version change (versioned markers keep it idempotent)
  • skills/<name>/SKILL.md — skill packs loaded on demand (aspnet-api-patterns, ef-core-patterns, testing-dotnet, clean-architecture, blazor-patterns, dotnet-project-init)

Authoring Rules

  • Hooks must be advisory by default. Exit 0 even on findings; emit guidance via additionalContext. Only block (non-zero exit) for hard safety violations. All hooks respect .planning/config.json hooks.* toggles and default-on when the file is absent.
  • Hook paths use ${CLAUDE_PLUGIN_ROOT}, never relative paths — the CWD at runtime is the consumer project, not this repo.
  • All hooks share _lib/config.js. When adding a new hook, import { hookEnabled } from there — don't re-implement config resolution.
  • dnp-commit-format skips heredoc commits. Claude Code's default multi-line commit workflow (-m "$(cat <<'EOF'...)") is deliberately excluded — the hook only validates plain -m "..." strings.
  • Commands are thin orchestrators. Heavy logic belongs in agents. A command file is the spec that Claude reads when the slash command fires; it should enumerate steps and which agents to spawn, not re-implement their work.
  • Agent frontmatter tools: is a whitelist. Adding a tool requires justification; prefer narrower scopes (Bash(dotnet:*)) over broad ones.
  • Model tier by agent role. Planning/architecture → opus; implementation/review → sonnet; mechanical checks (DI, NuGet audit, scaffolding) → sonnet at effort: low; deep read-only consults → fable.
  • Every agent and command declares effort:. Model tier sets capability, effort: (low|medium|high|xhigh|max) sets reasoning spend within it. Do NOT pair effort: with model: haiku — effort is unsupported on Haiku 4.5 and the field is silently dropped, which is why the mechanical agents run on sonnet at effort: low instead of haiku.
  • Agent prompts carry gotchas, not guardrails. Write what is specific to .NET and to this plugin — missing DI registration throwing at runtime, migrations needing their own step, the in-memory EF provider diverging from SQL Server. Do not add anti-rationalization tables, epistemic-gate checklists, "predict before you grep" protocols, or few-shot transcripts of ideal output: they re-encode judgment the model already has, and every added constraint is another chance to contradict a neighbouring one. Domain reference material (test tiers, mock fidelity, boundary coverage) belongs in a skills/ file the agent loads via skills:, not inlined in the prompt.
  • State a fact in exactly one place. Tool guidance goes in the tool or agent description; routing goes in this file; conventions go in a skill. A hook that re-injects a roster this file already carries pays for the same tokens on every call.
  • shell: is not used here. It only governs !-blocks in command frontmatter, and no dotnet-pilot command uses one — adding it would be dead config. If you introduce a !-block that shells out on Windows, add shell: powershell to that command.

Read the full file on GitHub · 218 lines

Changes

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.

  1. 7d ago First seen · 218 lines · 4,190 tokens per session scan B 19541da1f62e

Subscribe to this mod's changes

dotnet-pilot CLAUDE.md is an instructions file published in the GitHub repository zdanovichnick/dotnet-pilot (4 stars, last pushed 10d ago), licensed MIT. It adds 4,190 tokens to every session, about $0.0209 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens