dotnet-claude-kit AGENTS.md

Project instructions for routing coding tasks among specialist agents in a .NET toolkit, such as architecture, API design, database work, testing, security, performance, and deployment.

In plain words
What is it for?
Use them when maintaining or extending a .NET project that needs consistent agent selection, loading order, and coordination across development tasks.
Why use it?
They give the coding agent a shared map of which specialist should handle each kind of request and how those agents should coordinate.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/codewithmukesh/dotnet-claude-kit/agents-md
Clone the repo
git clone --depth 1 https://github.com/codewithmukesh/dotnet-claude-kit

Made for: Codex, OpenCode.

Per session 2,730 This file is loaded in full into every session.
When invoked 2,730 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.02730 $0.02730
Opus 5 $0.01365 $0.01365
Sonnet 5 $0.00546 $0.00546
Haiku 4.5 $0.00273 $0.00273

Measured 2d ago against content hash 29221d09ea08, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dotnet-claude-kit AGENTS.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 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.

AGENTS.md · 178 lines

How it starts

The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Routing & Orchestration

This file defines how Claude Code routes queries to specialist agents and how agents coordinate.

Agent Roster

Agent File Primary Domain
dotnet-architect agents/dotnet-architect.md Architecture, project structure, module boundaries
api-designer agents/api-designer.md Minimal APIs, OpenAPI, versioning, rate limiting
ef-core-specialist agents/ef-core-specialist.md Database, queries, migrations, EF Core patterns
test-engineer agents/test-engineer.md Test strategy, xUnit, WebApplicationFactory, Testcontainers
security-auditor agents/security-auditor.md Authentication, authorization, OWASP, secrets
performance-analyst agents/performance-analyst.md Benchmarks, memory, async patterns, caching
devops-engineer agents/devops-engineer.md Docker, CI/CD, Aspire, deployment
code-reviewer agents/code-reviewer.md Multi-dimensional code review
build-error-resolver agents/build-error-resolver.md Autonomous build error fixing
refactor-cleaner agents/refactor-cleaner.md Systematic dead code removal and cleanup

Routing Table

Match user intent to agent. When multiple agents could handle a query, the first match wins.

User Intent Pattern Primary Agent Support Agent
"set up project", "folder structure", "architecture" dotnet-architect
"add module", "split into modules", "bounded context" dotnet-architect
"create endpoint", "API route", "OpenAPI", "swagger" api-designer
"versioning", "rate limiting", "CORS" api-designer
"database", "migration", "query", "DbContext", "EF" ef-core-specialist
"write tests", "test strategy", "coverage" test-engineer
"WebApplicationFactory", "Testcontainers", "xUnit" test-engineer
"security", "authentication", "JWT", "OIDC", "authorize" security-auditor
"performance", "benchmark", "memory", "profiling" performance-analyst
"caching", "HybridCache", "output cache" performance-analyst
"Docker", "container", "CI/CD", "pipeline", "deploy" devops-engineer
"Aspire", "orchestration", "service discovery" devops-engineer
"review this code", "PR review", "code quality" code-reviewer
"choose architecture", "which architecture", "architecture decision" dotnet-architect
"scaffold feature", "create feature", "add endpoint", "generate feature" dotnet-architect api-designer, ef-core-specialist
"init project", "setup project", "new project", "generate CLAUDE.md" dotnet-architect
"health check", "analyze project", "project report" code-reviewer dotnet-architect
"review PR", "review changes", "code review", "PR review" code-reviewer
"add migration", "ef migration", "update packages", "upgrade nuget" ef-core-specialist
"conventions", "coding style", "detect patterns", "code consistency" code-reviewer
"add feature" (architecture-appropriate) dotnet-architect api-designer, ef-core-specialist
"refactor" code-reviewer dotnet-architect
"build errors", "fix build", "won't compile" build-error-resolver
"clean up", "dead code", "unused code", "de-sloppify" refactor-cleaner

Read the full file on GitHub · 178 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. 2d ago First seen · 178 lines · 2,730 tokens per session scan A 29221d09ea08

Subscribe to this mod's changes

dotnet-claude-kit AGENTS.md is an instructions file published in the GitHub repository codewithmukesh/dotnet-claude-kit (689 stars, last pushed 26d ago), licensed MIT. It adds 2,730 tokens to every session, about $0.0136 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.