novotnyllc

Routes .NET/C# requests to the correct domain skill and loads coding standards as baseline for all code paths. Determines whether the task needs API, UI, testing, devops, tooling, or debugging guidance based on prompt analysis and project signals, then…

Builds ASP.NET Core APIs, EF Core data access, gRPC, SignalR, and backend services with middleware, security (OAuth, JWT, OWASP), resilience, messaging, OpenAPI, .NET Aspire, Semantic Kernel, HybridCache, YARP reverse proxy, output caching, Office documents…

Baseline C# skill loaded for every .NET code path. Guides language patterns (records, pattern matching, primary constructors, C# 8-15), coding standards, async/await, DI, LINQ, serialization, domain modeling, concurrency, Roslyn analyzers, globalization,…

Debugs Windows and Linux/macOS applications (native, .NET/CLR, mixed-mode) with WinDbg MCP (crash dumps, !analyze, !syncblk, !dlk, !runaway, !dumpheap, !gcroot, BSOD), dotnet-dump, lldb with SOS, createdump, and container diagnostics (Docker, Kubernetes).…

Configures .NET CI/CD pipelines (GitHub Actions with setup-dotnet, NuGet cache, reusable workflows; Azure DevOps with DotNetCoreCLI, templates, multi-stage), containerization (multi-stage Dockerfiles, Compose, rootless), packaging (NuGet authoring, source…

Defines .NET test strategy and implementation patterns across xUnit v3 (Facts, Theories, fixtures, IAsyncLifetime), integration testing (WebApplicationFactory, Testcontainers), Aspire testing (DistributedApplicationTestingBuilder), snapshot testing (Verify,…

Manages .NET SDK installation (dotnet-install, workloads), project setup (.slnx, Directory.Build.props, CPM), MSBuild authoring, build optimization, performance (Span, ArrayPool, stackalloc), profiling (dotnet-counters, dotnet-trace), Native AOT/trimming, GC…

Builds .NET UI apps across Blazor (Server, WASM, Hybrid, Auto), MAUI (XAML, MVVM, Shell, Native AOT), Uno Platform (MVUX, Extensions, Toolkit), WPF (.NET 8+, Fluent theme), WinUI 3 (Windows App SDK, MSIX, Mica/Acrylic, adaptive layout), and WinForms…

Clean completed-session process residue, inspect macOS Codex app-server resources, reap exact snapshot-bound residue, or explicitly recycle one fully attested detached server.

Route one host-local software change or pull-request task through Railyard. Implementation/ship requests MUST dispatch an LFG delivery carrier before implementation or remote shipping; this coordinator never substitutes direct implementation for the LFG…

Oracle second-model review: bundle prompts/files, debug, refactor, design-check.

Comprehensive security and correctness audit of a branch's changes. Use for thermo nuclear, thermonuclear, or deep review requests, or branch/PR diff audits focused on bugs, breaking changes, security issues, devex regressions, and feature-gate leaks.

Orchestrate configured fleet/account delivery or complex objectives across independently resumable tasks, projects, hosts, pull requests, and dependencies while remaining available as the control task. Use when configured routing owns allocation, when the…

Resolve one bounded model, effort, transport, and budget decision through railyard/model-routing/v1. Use before model-specific task, subagent, provider-review, or steering actions.

Set up, extend, or diagnose the railyard delivery system: inventories what is installed, installs prerequisites (Compound Engineering, ponytail, roundhouse, agent-utilities, gh-stack, tmux/jq) with one consent, asks what the fleet hosts are and enrolls each…

Reconstruct how a chunk of agent work actually ran — which skills routed, what decided what, how many subagents fanned out on which models and why, retries, review rounds, and whether it matched the planned shape. Use when the user asks how a run worked, to…

Start and operate Codex Director. Use when the user mentions @codex-director, invokes $codex-director:codex-director, or asks to coordinate project work with Director threads.

Runs the Codex Director Browser ChatGPT Pro oracle lane for external second-opinion review through the signed-in browser. Use when a Director brief explicitly invokes $codex-director:director-browser-oracle for delegated oracle review, Pro-only comparison,…

Runs the Codex Director build lane for bounded implementation packets. Use when a Director worker brief explicitly invokes $codex-director:director-build for code, docs, configuration, tests, commits, or other implementation work that one worker can plan,…

Runs the Codex Director deep planning lane for durable implementation or architecture plans. Use when a Director worker brief explicitly invokes $codex-director:director-deep-plan for a plan document, migration strategy, architecture decision, decomposition,…

Runs the Codex Director dynamic workflow bridge for complex task-level packetization and durable orchestration artifacts. Use when a Director brief explicitly invokes $codex-director:director-dynamic-workflow for approvals, packets, multi-repo or multi-worker…

Runs the Codex Director investigate/research lane for read-only diagnosis and evidence gathering. Use when a Director worker brief explicitly invokes $codex-director:director-investigate for repo or docs reconnaissance, failure diagnosis, external fact…

Runs the Codex Director optimization lane for performance, cost, memory, bundle size, throughput, or latency work. Use when a Director worker brief explicitly invokes $codex-director:director-optimize for measurement-led improvement, bottleneck diagnosis,…

Runs the Codex Director worker orchestration lane for multi-item or ambiguous bounded assignments. Use when a Director worker brief explicitly invokes $codex-director:director-orchestrate, or when a worker must coordinate subwork, helper/subagent lanes,…

Install, update, or remove a resilient stdio MCP shim in front of a desktop app's HTTP MCP server (Autodesk Fusion, Figma, or any other app/URL), for Claude Code and/or Codex. Use when the user wants to "set up MCP for" a desktop app, register an MCP server…

Plan and explicitly apply package updates across Homebrew, APT, and winget machines. Use for fleet patching, outdated-package reports, manager-specific updates, package drift, or post-update verification.

Inventory and reconcile Codex and Claude runtimes, safe settings, plugins, standalone skills, skills-cli provenance, JSM-managed skills, and logical capabilities across machines. Use when agent tooling, Remote Control defaults, models, updates, or skills are…

Add a host to the fleet or remove one, end to end: config entry, SSH reachability, SSH-certificate identity enrollment or revocation, privilege-broker enrollment where wanted, target prerequisites (agent harnesses, plugins, tmux/jq), and readiness…

Audit and deliberately reconcile configured authentication artifacts across machines without exposing secrets. Use for Codex auth.json, xurl auth.yml, GitHub or tool sessions, matching credential hashes, native auth health, reauthentication, or encrypted…

Inventory and compare a configured fleet of macOS, Linux, WSL, and Windows machines. Use for packages, agent runtimes, allowlisted Claude/Codex settings, plugins, standalone skills and provenance, auth-file or native session status, projects, startup tasks,…

Assess and reconcile project, agent, plugin, skill, authentication, and host readiness across configured machines. Use before cross-host task placement, when fleet capabilities may be missing or inconsistent, or when a user asks whether machines are ready for…

Operate and diagnose configured remote Macs over Tailscale or SSH, with tmux and GUI fallback. Use when a fleet Mac needs inspection, repair, service checks, or remote automation.

Use for Codex multi_agent_v2 subagent orchestration: delegation decisions, spawn_agent prompts, custom agent_type roles, reasoning_effort choices, safe parallelism, wait/list/send/followup/close handling, commit discipline, receipts, and root-thread…

Build with RepoPrompt MCP tools context builder plan → implement

Deep planning workflow using RepoPrompt MCP tools: map seams, draft, critique, polish — produces a ready-to-execute plan document

Deep investigation with RepoPrompt MCP tools: tools gather evidence, follow-up reasoning synthesizes selected context

Iterative performance optimization loop using RepoPrompt MCP tools: instrument with debug-only metrics, establish a baseline, then plan → delegate one optimize+harden cycle → re-measure → ask oracle for next plan, looping until the oracle is satisfied or the…

Export a ChatGPT-ready Question / Plan / Review prompt using RepoPrompt MCP tools

Plan, decompose, and delegate complex tasks across multiple agents using RepoPrompt MCP tools

Refactoring assistant using RepoPrompt MCP tools to analyze and improve code organization

Control the user's existing signed-in Chrome: native browser tool first, with direct DevTools attachment only as a last fallback.

CLI UX/spec: args, flags, help, output, errors, config, dry-run.

Inspect, compare, and deliberately reconcile chezmoi source and live-state drift across configured machines. Use for chezmoi status, diff, pull, add, apply, source-repository drift, or post-apply verification.

Use when creating, designing, editing, repairing, inspecting, or validating a CAD model, Autodesk Fusion model/design, dimensioned physical 3D model, or 3D-printable part. Prefer this skill automatically for requests to make a Fusion or CAD model and for…

1Password/op: service-account first, targeted secret read/store/inject, tmux.

macOS OneDrive File Provider diagnosis and bounded FPCK repair for a real ~/Documents/Codex directory. Use for high fileproviderd, suggestd, or OneDrive CPU; provider churn; File Provider reconciliation errors; or a safe fileproviderctl repair that must…

Agent skill audit: live budget, usage, duplicates, compact descriptions.

Use the official UniFi Network Integration API for managed UniFi console automation, especially firewall policies, policy ordering, zones, networks, clients, devices, DNS policies, traffic matching lists, and other Network app configuration. Use when the user…

Run any command, build, or test in a disposable macOS virtual machine (Tart). Use whenever the user wants work run in a VM, sandbox, container, clean room, or fresh/isolated/ephemeral/disposable environment, off-host, "not on my machine", or without polluting…

Browser automation setup and routing for Chrome and Microsoft Edge stable, beta, dev, and canary. Use when the user mentions @browser-control, @chrome-* or @edge-*, or asks for Chrome Dev, Chrome Canary, Edge Dev, or Edge Canary.

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times,…