Real-time command explanations, risk assessment, and audit trails for developers who work with AI-generated and unfamiliar terminal commands.
Start for free | Read the docs
Works with any terminal. No lock-in to a specific editor or shell.
Every command is explained instantly as you type. Shell hooks capture your commands and return structured breakdowns in milliseconds.
Every command is scored as safe, caution, or danger. Destructive operations and risky permission changes are flagged before execution.
Your command history syncs across devices. Search, filter by risk, and maintain a complete audit trail for compliance.
Terminall works entirely locally by default. The CLI caches explanations in local SQLite.
Cloud sync is opt-in. PII redaction strips secrets before any API call.
For maximum privacy, use local_only: true with a local Ollama model.
Terminall uses preexec hooks (same mechanism as atuin, starship, direnv) to capture commands before they execute. Supported: zsh, bash. Fish shell support planned. Works on macOS and Linux. Windows via WSL is experimental.
One command to install. One command to activate.
Or via npm:
Start free, upgrade when you need more. No surprise fees.
20 cloud commands/day. Unlimited local commands. 7-day history.
Unlimited cloud & local commands. 30-day history. Sonnet on-demand.
Unlimited commands. 90-day history. Team workspace. Audit logs.
Free tier includes 20 cloud explanations per day. No credit card required. Get started free | View full pricing
terminall init — Initialize shell hooks and create default config
terminall explain <cmd> — Explain a command without running it
terminall tui — Launch split-pane TUI mode
terminall watch — Tail-style live feed of recent commands
terminall login — Authenticate with Terminall cloud
terminall sync — Manually sync local history to cloud
terminall history — Show recent command history
terminall cache clear — Clear the local explanation cache