napisani

How to declaratively add, change, or remove AI coding agent assets — skills, MCP servers, plugin/capability installs, RTK hooks, shared instructions — in this home-manager repo's mods/agents/*.nix module. Use this whenever the user asks to add a skill, add an MCP server, install a plugin for Claude/Codex/OpenCode/Pi, gate something to a specific machine, update a pinned skill, or asks why removing something from mods/agents/*.nix didn't actually remove it after a rebuild. Also use this before writing any new code under mods/agents/ or mods/dotfiles/agents/scripts/, even if the user doesn't name this skill directly — this architecture has specific, non-obvious rules (see "The one rule that matters" and "The three layers") that are easy to violate by copying an old pattern.

Resolve conflicts after Stackman stops during a rebase. Use when stackman sync reports rebase conflicts, when the user asks to continue a Stackman rebase, or when Git is in a conflicted rebase caused by Stackman rebasing a branch onto its parent/upstream branch.

This skill should be used when the user asks about "neovim config", "nvim setup", "vim best practices", "neovim patterns", "modern neovim", "lua configuration", "neovim 0.10", or wants guidance on configuring Neovim following current standards and conventions.

Use when the user wants to handle GitHub pull request feedback, review comments, reviewer suggestions, pasted PR comments, or a PR URL with comments that need triage, code changes, or drafted reviewer replies.

Use when the user explicitly asks to store, create, organize, read from obsidian notes or obsidian vault.

Guide for this Neovim configuration -- a modular Lua-based IDE rooted at mods/dotfiles/nvim/. Use when configuring plugins, adding keybindings, setting up LSP servers, debugging, or extending the config. Covers lazy.nvim, plugin_registry, snacks.nvim pickers, blink.cmp completion, dual LSP architecture, DAP debugging, and which-key v3 aggregation.

Write or update BEHAVIOR.md files that document tool/plugin workflows as portable behavior contracts. Use when the user wants to document keybindings, workflow behaviors, tool conventions, or create per-tool behavior specs for Neovim, tmux, bash, or other keyboard-driven tools. Also use when the user wants to capture "what this tool should do" independently of the current plugin implementation, so that plugins can be swapped without losing workflow preferences.

Implement Neovim plugins and configurations with TDD. Invoke for lua-language implementation tasks.

Use when adding, running, or debugging proctmux end-to-end integration tests, especially PTY/TUI behavior, terminal snapshots, unified/client modes, timing flakes, VT100 emulator limits, or files under tests/e2e and internal/testharness/e2e.

This skill should be used when designing terminal user interfaces, creating TUI layouts, choosing TUI color schemes, implementing keyboard navigation, building terminal dashboards, or working with any TUI framework. Activates on mentions of TUI design, terminal UI, Ratatui layout, Ink components, Textual widgets, Bubbletea views, terminal color palette, keybinding design, panel layout, split panes, terminal dashboard, box-drawing characters, sparklines, progress bars, modal dialogs, focus management, or terminal accessibility.

Use this skill whenever the user wants to create, edit, explain, validate, or troubleshoot a proctmux.yaml/procmux.yaml file, add or change proctmux processes, configure lifecycle behavior such as stop signals/on_kill/autostart, customize unified-mode layout/keybindings/style, or enable Makefile/package.json process discovery. Use it even when the user says this casually, such as "add my dev server to proctmux", "fix this proctmux config", "write a config for these services", or "what YAML option controls focus".

Automate terminal UI (TUI) apps with agent-tui for testing, inspection, demos, and scripted interactions. Use when automating CLI/TUI flows, regression testing terminal apps, verifying interactive behavior, or extracting structured text from terminal UIs. Also use when asked what agent-tui is, how it works, or to demo it. Do not use for web browsers, GUI apps, or non-terminal interfaces.

Harness-first workflow for evaluating and improving terminal UIs through real interaction. Use this whenever the user wants to review a TUI experience, compare before/after behavior, identify UX friction, validate navigation or filtering flows, inspect visual stability, or turn observed interaction problems into concrete follow-up tests and improvement recommendations. Prefer this skill any time a TUI should be exercised through a reproducible PTY/session harness instead of only reading code.

Writes a reviewable code walkthrough for vantage.nvim at .vantage/walkthrough.json — an ordered list of code pointers (file + line), each with a short annotation, distilled from the current session. Vantage loads it into a Neovim quickfix list and shows each annotation inline as the developer visits the line. Use this whenever the developer wants to review, walk through, or annotate specific lines of code in their editor based on this conversation — for example "show me the spots we changed in nvim", "mark up these files for review", "turn this into a walkthrough", or "point Vantage at the lines we discussed" — even if they never say the word "walkthrough". Prefer this over vantage-distill-session when the developer wants pointers to specific lines to review, rather than a general prose snapshot of the session.

Distills the current adjacent-agent session into Vantage's workspace context artifact at .vantage/agent-context.md. Use when the user wants Vantage, vantage.nvim, or Neovim to use the current agent session context, or asks to update/distill/snapshot Vantage agent context.

Run agentic PTY smoke tests for scute (shell integration, output channels, and TUI) with scripted scenarios and log capture.