getv — Universal .env Variable Manager

We increasingly work with APIs, especially in the context of LLMs, constantly needing to provide new keys and then copy them to other tools, .env files. What if when copying to the clipboard, they were automatically saved to .env from which various applications like ollama or libraries like liteLLM could retrieve the specified API TOKEN?

• copy an API KEY in your browser
• run: getv grab
• use multiple keys directly: getv exec llm groq -- python app.py

Read, write, encrypt, and delegate environment variables across services and devices.

Copy to the clipboard and run getv grab to detect and save the API key

$ getv grab

Detected:  groq (GROQ_API_KEY)
Key:       gsk_Y1xV...TNpA
Source:    Prefix match
Domain:    console.groq.com
Category:  llm
Profile:   ~/.getv/llm/groq.env

Saved to /home/tom/.getv/llm/groq.env

Usage:
  getv get llm groq GROQ_API_KEY
  getv exec llm groq -- python app.py

without any plugins, managers, or integrations...

Clipboard → .env Auto-Detection No other CLI tool has this feature. Copy any API key and run getv grab — it auto-detects the provider, saves to the right profile, and shows usage commands. Perfect for rapid RPi + LLM development workflow.

$ getv grab              # copy API key → auto-detect provider → save
$ getv exec llm groq -- python app.py   # run with profile injected
$ getv ssh rpi3 "uname -a"              # SSH using stored profile

Why getv?

• Clipboard → .env — getv grab auto-detects 19 API key prefixes from clipboard
• Built-in integrations — SSH, curl, Docker, LiteLLM, Ollama, Pydantic — no plugins
• Smart profiles — organize by category (llm/, devices/, tokens/) with per-app defaults
• One-liner power — process substitution, pipes, shell eval

Install

pip install getv                   # core
pip install "getv[crypto]"         # + encryption (Fernet)
pip install "getv[all]"            # everything

Quick Start

# Save a profile
getv set llm groq LLM_MODEL=groq/llama-3.3-70b-versatile GROQ_API_KEY=gsk_xxx

# Read a variable
getv get llm groq LLM_MODEL

# List profiles (secrets masked)
getv list llm

# Run with env injected
getv exec llm groq -- python my_script.py

# Auto-detect API key from clipboard
getv grab

# Export formats
getv export llm groq --format json
getv export llm groq --format shell
getv export llm groq --format docker

Python API

from getv import EnvStore, ProfileManager

# Single .env file
store = EnvStore("~/.myapp/.env")
store.set("DB_HOST", "localhost").set("DB_PORT", "5432").save()

# Named profiles
pm = ProfileManager("~/.getv")
pm.add_category("llm", required_keys=["LLM_MODEL"])
pm.set("llm", "groq", {"LLM_MODEL": "groq/llama-3.3-70b-versatile", "GROQ_API_KEY": "gsk_xxx"})

# Merge profiles
cfg = pm.merge_profiles({"APP_NAME": "myapp"}, llm="groq", devices="rpi3")

Profile Directory

~/.getv/
├── .fernet.key            ← encryption key (chmod 600)
├── defaults/              ← per-app profile selections
│   ├── fixpi.conf         → llm=groq, devices=rpi3
│   └── prellm.conf        → llm=openrouter
├── devices/
│   ├── rpi3.env
│   └── rpi4-prod.env
├── llm/
│   ├── groq.env
│   └── openrouter.env
└── tokens/
    └── github.env

CLI Reference

Documentation

Environment Variables

Adopted by

• fixpi — SSH + LLM diagnostic agent
• prellm — LLM preprocessing proxy
• code2logic — Code analysis engine
• amen — Intent-iterative AI gateway
• marksync — Markdown sync server
• curllm — LLM-powered web automation

Development

git clone https://github.com/wronai/getv.git
cd getv
python -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
pytest  # 190 tests

License

Apache License 2.0 - see LICENSE for details.

Author

Created by Tom Sapletta - tom@sapletta.com