Your code is never stored. GDPR-compliant processing with immediate data deletion.
View our security standardsLurus Code CLI
Full AI coding agent in your terminal. Interactive chat, batch processing, CI security scans, code review automation, and shell completions. GDPR-compliant.
Interactive Chat
Start a session with `lurus chat` and work with the AI right in the terminal.
Batch Processing
Run many tasks with `lurus batch`, ideal for scripts and pipelines.
CI Security
Integrate security scans with `security-ci` into your build and release pipelines.
CI Code Review
Automated reviews with `code-review-ci` for pull requests and merge queues.
Shell Completions
Tab completion for bash, zsh, and fish, navigate faster and type less.
Plugin Management
Install and manage extensions with `lurus plugin`, wired into the ecosystem.
Get Started in 60 Seconds
Install the CLI globally via npm, authenticate once, and start your first AI coding session immediately.
# Install globally npm install -g @scramble-cloud/lurus-code-cli # Authenticate (browser, email, or API key) lurus login # Start interactive session lurus chat # Or send a single prompt without opening a session lurus "Explain the architecture of this project" # Continue last session for this project lurus chat --continue
Batch Processing
Process multiple AI tasks from a file sequentially — perfect for scripted workflows, documentation generation, or automated refactoring across a codebase.
Example Prompt File
# Prompt file (prompts.txt) # Separate prompts with --- Analyze src/auth.service.ts and explain the authentication flow --- Write unit tests for UserService --- Generate a README for the api/ module
Commands
lurus batch prompts.txt Run all prompts sequentially lurus batch prompts.txt -o json JSON output with token counts lurus batch prompts.txt --continue-on-error Skip failed prompts and continue lurus batch prompts.txt -m haiku Use a faster model for simple tasks CI/CD Integration
Two dedicated headless commands for CI pipelines: security-ci and code-review-ci. Both return standard exit codes and support GitHub Actions output variables.
- name: AI Security Scan
run: lurus security-ci --diff --fail-on high --format sarif
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} - name: AI Code Review
run: lurus code-review-ci --pr-comments --verdict --fail-on high
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} Shell Tab Completions
Enable Tab completion for all lurus commands, subcommands, and flags in your preferred shell.
echo 'eval "$(lurus completions bash)"' >> ~/.bashrc echo 'eval "$(lurus completions zsh)"' >> ~/.zshrc lurus completions fish > ~/.config/fish/completions/lurus.fish Permission Modes
Control how much autonomy the agent has. All modes can be set per session or persisted in your project configuration.
| Mode | Behavior | Use Case |
|---|---|---|
default | Prompts for confirmation before every tool use | Day-to-day development |
acceptEdits | Auto-approves file edits, prompts for shell commands | Trusting the agent with code changes |
plan | Read-only — no file writes or shell commands | Architecture analysis, code exploration |
bypassPermissions | Grants all permissions automatically | CI/CD pipelines only (use with --dangerously-skip-permissions) |
Ready to code smarter?
Get started and use the Lurus Code CLI with every command and integration.
Get started