Rules

AI coding agent instructions for project conventions and guidelines.

2. Reliability and safety 3. Cross-platform correctness 4. Minimal, composable design 5. Backward compatibility 6. Performance on constrained systems 7. Dependency and binary size discipline

Notes for future LLM sessions working in this repo. Read this before spelunking.

Guidance for AI coding agents working in this repository. Human contributors should also read [docs/contributing.md](docs/contributing.md), [docs/development.md](docs/development.md), and [docs/testing.md](docs/testing.md), which are the source of truth.

Read design.md to get an overview of huxtable's architecture. Read agent-notes.md to read the notes of previous llms.

This project uses **Nix flakes** for all builds and development tooling. The cloud agent VM has Nix installed via the Dockerfile, but the nix-daemon must be running for any Nix command to work.

This repository mixes several deliverables that move at different speeds. To keep things tidy, we split the work into three "agents" that each own one slice of the Rust on Nails experience. Use this guide to decide where changes belong and how to validate them before opening a PR.

All new commands must use the struct-based pattern. **Do not copy the old flat `var cmd = &cobra.Command{...}` commands** - they exist in `cmd/storage/`, `cmd/dns/`, `cmd/dbaas/` and are legacy. Read this file before writing any new command:

AI coding agent instructions for project conventions and guidelines.

> **Methodology:** [Trace-First Workflow (TFW)](https://github.com/saubakirov/trace-first-starter) v2.0.0-dirty > **Framework files:** `.tfw/` · **Conventions:** `.tfw/conventions.md` · **Glossary:** `.tfw/glossary.md`

Monorepo with three packages that are NOT linked via workspaces — each has its own `node_modules`:

> **⚠️ This file has been migrated to the modern structure** > This project now uses Cursor's Rules, Skills, and Subagents structure for better AI agent support: > - **Rules** (`.cursor/rules/`): Always-applied project standards

Refer to the following files for mandatory instructions: If there are conflicting rules in this or the documents below, prompt for clarification.

SCAM (Security Comprehension Awareness Measure) is a CLI benchmark that evaluates the safety of AI agents with tool access. It runs multi-turn conversations where an AI agent uses tools (inbox, browser, forms, credential vault) and must proactively protect the user from threats without being told to look for them.

> Pure Go 2D graphics library. Smart rasterizer, GPU acceleration, text rendering, SVG paths.

211 production GTM skills for AI agents. This repository follows the Agent Skills open specification: portable skill folders with SKILL.md plus optional scripts/, references/, templates/, and assets/.

This document defines the operational agents that are implemented in the Hakumai macOS app today. It is intended for contributors who add features, fix bugs, or design new automation around live comment ingestion and moderation support.

This repo runs Chrome's on-device Gemini Nano model behind a local OpenAI-compatible API. If you're an agent that wants to use it as a model backend (not modify the repo itself), this is what you need.

This cookbook manages the xinetd extended internet services daemon through custom resources: `xinetd_install`, `xinetd_config`, `xinetd_service`, and `xinetd_builtin_services`.

These instructions are for AI assistants working in this project. Use the `/trellis:start` command when starting a new session to: Use `@/.trellis/` to learn: If you're using Codex, project-scoped helpers may also live in:

> Operating manual for LLM agents working with this DQMC codebase.

This repo is not expected, for now, to be imported as a dependency; treat exported internals as pi-vim-local unless documented otherwise. For every new or changed Vim-like feature, add curated nvim parity coverage in `test/nvim-parity*.ts` unless the behavior is intentionally not Vim-compatible. If it is an intentional divergence, make that explicit in tests and documentation.