Dev logo

Dev

OrganizationPopular
slopus
dev

Local development guide for the Happy monorepo. How to build, install, test, and run the CLI, server, mobile app, and desktop (Tauri) locally. Use when the user types /dev, asks how to "build", "start dev", "install locally", or "run the ___ package".

Overview

Publisherslopus
Repositoryhappy
Skill namedev
Stars
24.1K
Forks
2.1K
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by slopus on GitHub. Read the source before you install it.

Installation

Install the Dev AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/slopus/happy/tree/main/.agents/skills/dev
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/slopus/happy.git /tmp/happy
mkdir -p .claude/skills
cp -r /tmp/happy/.agents/skills/dev .claude/skills/dev
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Dev in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Dev on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Dev is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

/dev - Local Development

Happy is a pnpm monorepo. Everything uses pnpm workspaces — do not use npm or yarn directly.

First-time setup

bash
pnpm install                       # installs deps for every package
pnpm --filter happy cli:install    # builds happy-cli + links it as the global `happy` binary

cli:install replaces whatever happy is on your PATH (npm-installed or not) with a symlink to packages/happy-cli/. Daemon is restarted as part of the script. Uses ~/.happy/ — same as production.

To undo: npm unlink -g happy && npm i -g happy@latest.

Packages

packages/happy-cli     # the `happy` CLI and daemon, published to npm
packages/happy-server  # Node + Prisma server, deployed via TeamCity
packages/happy-app     # Expo app: iOS, Android, web, Tauri desktop
packages/happy-agent   # agent runtime
packages/happy-wire    # shared Zod schemas + wire types

happy-cli

packages/happy-cli
scripts in package.json:
  typecheck      # tsc --noEmit
  build          # rm -rf dist && tsc --noEmit && pkgroll
  test           # build + vitest run
  cli:install    # build + stop daemon + npm link + start daemon
  prepublishOnly # pnpm test (runs build inside test)
  postinstall    # unpacks difft + rg binaries into tools/unpacked/

Work loop:

bash
pnpm --filter happy cli:install   # rebuild + relink + restart daemon
happy daemon status               # confirm your build is running
happy doctor                      # list all happy processes
tail -f ~/.happy/logs/$(ls -t ~/.happy/logs/ | head -1)

Run a single test file quickly:

bash
pnpm --filter happy exec vitest run src/path/to/file.test.ts

Unit-only (fast, ~1 min):

bash
pnpm --filter happy exec vitest run --project unit

Integration tests hit real APIs and are flaky — run on demand, never in the release gate.

Dev data sandbox (optional)

happy reads HAPPY_HOME_DIR to override ~/.happy/. To run two versions side-by-side without touching your prod auth:

bash
HAPPY_HOME_DIR=~/.happy-dev happy daemon start
HAPPY_HOME_DIR=~/.happy-dev happy auth

Point at a local server the same way:

bash
HAPPY_SERVER_URL=http://localhost:3005 happy daemon start

happy-server

bash
pnpm --filter happy-server standalone:dev   # localhost:3005, embedded PGlite, no Docker

App auto-reloads on source changes. Point the CLI or the Expo app at it with HAPPY_SERVER_URL=http://localhost:3005 / EXPO_PUBLIC_HAPPY_SERVER_URL=....

happy-app (Expo)

bash
pnpm --filter happy-app start           # expo start (Metro bundler)
pnpm --filter happy-app ios:dev         # iOS simulator, development variant
pnpm --filter happy-app android:dev
pnpm --filter happy-app web             # web build, served locally
pnpm --filter happy-app tauri:dev       # macOS desktop app

Variants:

development    com.slopus.happy.dev       # hot reload, internal
preview        com.slopus.happy.preview   # OTA / beta testing
production     com.ex3ndr.happy           # App Store

Rebuild and reinstall the desktop .app

When the user asks to "rebuild the desktop app", "kill the running one and reinstall", or anything in that shape — do all four steps in order, do not stop after building.

Variants → product name → build script:

production    Happy.app           pnpm --filter happy-app tauri:build:production
preview       Happy (preview).app pnpm --filter happy-app tauri:build:preview
dev           Happy (dev).app     pnpm --filter happy-app tauri:build:dev

Build output for all variants:

packages/happy-app/src-tauri/target/release/bundle/macos/<ProductName>.app

If the variant is ambiguous, check what's running with ps aux | grep "/Applications/.*Happy" | grep -v grep and match. Production is the default.

Steps (substitute $NAME with the product name, e.g. Happy or Happy (dev)):

bash
# 1. build (slow: ~3–10 min, expo web export then cargo release build)
pnpm --filter happy-app tauri:build:production

# 2. quit the running app gracefully (no-op if not running)
osascript -e 'tell application "$NAME" to quit' || true

# 3. replace the installed bundle
rm -rf "/Applications/$NAME.app"
cp -R "packages/happy-app/src-tauri/target/release/bundle/macos/$NAME.app" /Applications/

# 4. relaunch
open -a "$NAME"

Notes:

  • Run the build in the background (run_in_background: true on Bash) and poll the output file. It prints Finished \release` profile` near the end.
  • osascript ... to quit is graceful — it gives the app a chance to flush state. Only fall back to pkill -f "/Applications/$NAME.app/Contents/MacOS/app" if the quit hangs.
  • Do NOT skip the rm -rf before cp — cp -R over an existing .app merges directories and leaves stale files.
  • If macOS Gatekeeper complains on relaunch, xattr -dr com.apple.quarantine "/Applications/$NAME.app" clears it. Local builds are unsigned.

happy-app-logs (remote log receiver)

bash
pnpm --filter happy-app-logs dev       # starts on http://0.0.0.0:8787

Receives POST requests to /logs from the mobile app's patched console (see consoleLogging.ts). Logs to stdout and ~/.happy/app-logs/<timestamp>.log.

To connect: set the log server URL in the app's dev settings to http://<LAN_IP>:8787. The app's consoleLogging.ts sends all console.log/warn/error to this endpoint when configured.

Console output must be enabled in the app (dev/preview variants default on, production defaults off, togglable from the dev settings screen).

Cross-cutting

  • Hoisted deps: pnpm hoists node_modules to the repo root. packages/*/node_modules/ is mostly empty. Node's resolution walks up, so imports work transparently.
  • Workspace deps: "@slopus/happy-wire": "workspace:*" resolves to packages/happy-wire/ — edits are picked up live.
  • $npm_execpath: legacy; happy-cli uses pnpm literally. Windows cmd.exe doesn't expand $VAR.
  • Build before tests: tests spawn the built CLI binary (for daemon integration), so pnpm test runs build first. Do not remove.

Releasing

Do not publish by hand. Use /release — it handles npm publish, git tags, GitHub releases, and the smoke check.

Troubleshooting

happy: command not found     → pnpm --filter happy cli:install
daemon won't start           → happy daemon stop; rm ~/.happy/daemon.state.json.lock; happy daemon start
wrong `happy` version        → which happy && ls -la $(which happy) — confirms where it resolves to
tools/unpacked missing       → pnpm install (postinstall re-extracts)
stale deps after branch swap → pnpm install (pnpm is picky about lockfile drift)

Rules

  • Never use npm install or yarn install — only pnpm.
  • Never add a dev / cli tsx-based script back to happy-cli. The build step is not optional — daemon spawns the built binary and would desync.
  • Never bring back release-it. Releases go through /release.
  • Never introduce ~/.happy-dev as a default. It exists as an opt-in via HAPPY_HOME_DIR, nothing more.

Frequently asked questions

What does the Dev AI skill do?

Local development guide for the Happy monorepo. How to build, install, test, and run the CLI, server, mobile app, and desktop (Tauri) locally. Use when the user types /dev, asks how to "build", "start dev", "install locally", or "run the ___ package".

Why use Dev on TypingMind?

Because you install it once and use it with any model. Dev is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Dev in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/slopus/happy/tree/main/.agents/skills/dev. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Dev?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Dev?

As many as you like. As long as a model supports skills, you can use Dev with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Dev AI skill free?

Yes. It is published on GitHub by slopus under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇