Add local, in-process LLM inference to an existing .NET project with Overfit — no Python, no Ollama, no cloud. Use when the user wants to run a private/local LLM inside their .NET app, load a GGUF model in C#, add chat/RAG/embeddings on the CPU, replace an OpenAI/Ollama/Azure call with an on-device model, or expose a local model as a Microsoft.Extensions.AI IChatClient. For a brand-new app, prefer the `dotnet new overfit-chat` template instead.
Build and run an eval harness for an agent skill or prompt LOCALLY with Overfit — deterministic (seeded/greedy), offline, zero API cost, with schema-guaranteed rubric grading. Use when asked to test/evaluate/score a skill or prompt, catch prompt regressions, or measure whether a prompt change helped.
Spec-driven development for the Overfit engine. Use when starting a new feature, model/op/kernel/loader/runtime change, or any change touching multiple files — writes an Overfit-native spec (architecture path, AOT / zero-alloc / parity gates, git-read-only boundary) and a gated plan BEFORE coding. Prefer this over the generic spec-driven-development skill inside this repo.
Creates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when requirements are unclear, ambiguous, or only exist as a vague idea.
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Claude Code skill for checking and updating NuGet packages in a .NET 10 solution using native CLI commands.
Identifies high-quality leads for your product or service by analyzing your business, searching for target companies, and providing actionable contact strategies. Perfect for sales, business development, and marketing professionals.