perky — CLI Workspace Launcher & AI Assistant

perky is a CLI that combines workspace automation and AI assistance for developers. It lets you name projects, start services, open tools, and ask or explain code using AI providers (Gemini, OpenAI, Ollama).

Status: Active development — core commands implemented and tested.

Highlights (recent updates):

• New perky start orchestration improvements: better concurrent service handling and detached mode.
• perky summarize now saves local JSON summaries and offers optional AI insights.
• Improved AI provider abstraction and streaming support (see src/ai/).
• Expanded project preset config and perky config helpers (add/list/remove projects).

Quick list of commands

• perky init — interactive setup for AI provider, model, editor, and optional project registration
• perky ask — AI Q&A (supports streaming and clipboard copy)
• perky explain — AI file explanations with --detail and --section
• perky summarize — summarize logs; outputs JSON summary and optional AI insights
• perky time — show local time (--live for continuous update)
• perky open — open editor, browser URLs, file explorer, terminal
• perky app <name> — open an installed app (alias: perky <app>)
• perky start — start configured services (attached or detached)
• perky go — normalize/open/copy/print URLs quickly
• perky shutdown / perky restart — system power actions (--delay available)
• perky update — update perky to latest published version
• perky config — set/get/list/add-project/remove-project/edit global config

Install

Using npm (published package name: perky):

npm i perky
# or
npm i -g perky

Local development:

git clone https://github.com/ArushKhasru/perky.git
cd perky
npm install
npm link

Run:

perky --help
# or
node cli.js --help

Configuration

Global config: ~/.perky/config.json — stores AI defaults and project presets. Project-local config: .perky.json in project root.

Environment variables supported:

• GEMINI_API_KEY
• OPENAI_API_KEY

perky init can persist keys into a local .env file.

Example global config structure is in config/defaults.json.

AI Providers

Supported providers:

• Gemini — uses GEMINI_API_KEY
• OpenAI — uses OPENAI_API_KEY
• Ollama — local server at http://localhost:11434

AI behavior and prompt templates are in src/ai/prompt.js and provider adapters in src/ai/provider.js.

Project presets

Register projects with perky config add-project <name> --path <path> and include optional editor, browser, and services entries so perky start and perky open work seamlessly.

Development & Testing

Run the test suite with:

npm test

Tests are under the tests/ directory and cover src/ai, src/commands, src/utils, and src/workspace.

Contributing

• Open issues or PRs at: https://github.com/ArushKhasru/perky
• Follow repo coding conventions (ESM, minimal changes, focused PRs)

License

ISC