---
title: "Commands Reference"
description: "All slash commands available in Cody CLI — code review, agents, workflows, configuration, and more."
canonical: "https://docs.ainative.studio/docs/cody-cli/commands"
last-updated: "2026-10-03T21:47:10.764Z"
---

# Commands Reference

Source: https://docs.ainative.studio/docs/cody-cli/commands

> All slash commands available in Cody CLI — code review, agents, workflows, configuration, and more.

# Commands Reference

Cody CLI provides 30+ slash commands. Type `/` in an interactive session to see the full list with autocomplete.

## Code Quality

| Command | Description |
|---------|-------------|
| `/code-review` | Review staged or last-commit changes for bugs, edge cases, and missing error handling |
| `/security-review` | OWASP Top 10 vulnerability scan — SQL injection, XSS, command injection, path traversal, hardcoded secrets, SSRF, insecure crypto, missing auth, insecure deserialization, security misconfiguration |
| `/pr-review [number\|url]` | Review a pull request for bugs, security issues, and performance |
| `/pr-fix [number]` | Fetch PR review comments and auto-generate code fixes |

## Agents & Automation

| Command | Description |
|---------|-------------|
| `/goal <condition>` | Work autonomously until a condition is met (max 20 turns) |
| `/bg <task>` | Delegate a task to a background agent — continue working in the foreground |
| `/auto on\|off\|status` | Toggle auto mode — safe actions run without prompting, dangerous ones still require approval |
| `/workflow run <file.yml>` | Execute a YAML-defined multi-step agent pipeline |
| `/workflow list` | List available workflows in `.cody/workflows/` |
| `/workflow init` | Create a sample workflow template |
| `/agent-view` | List all running, completed, and failed agents with status, duration, and token usage |
| `/agent-view <id>` | Inspect a specific agent |
| `/agent-view kill <id>` | Stop a running agent |
| `/loop N <instruction>` | Repeat an instruction N times |
| `/loop until <condition>` | Loop until a condition is met (max 20 iterations) |

## Session

| Command | Description |
|---------|-------------|
| `/recap [minutes\|all]` | Show what happened in the last N minutes (default: 5) |
| `/rewind` | Resume conversation from before `/clear` |
| `/usage` | Show token and cost tracking for the current session |
| `/monitor start <path>` | Watch files/directories for changes |
| `/monitor stop` | Stop file watching |
| `/monitor status` | Report accumulated changes since last check |
| `/notifications on\|off\|test` | Control mobile push notifications |
| `/compact` | Compress conversation context to free up tokens |
| `/clear` | Clear the conversation |

## Configuration

| Command | Description |
|---------|-------------|
| `/config [key=value]` | Set a config value inline (e.g., `/config model=kimi-k2`) or open the interactive settings panel |
| `/model` | Interactive model picker — filtered by your subscription tier |
| `/theme` | Switch UI theme |
| `/fast` | Toggle fast mode (same model, faster output) |

## Learning & Help

| Command | Description |
|---------|-------------|
| `/powerup [lesson]` | Interactive tutorials — 6 lessons covering tools, agents, MCP, memory, skills, and code review |
| `/help` | Show help |

## MCP & Skills

| Command | Description |
|---------|-------------|
| `/mcp-login <server>` | Authenticate with an MCP server via OAuth |
| `/mcp-logout <server>` | Clear stored MCP credentials |
| `/skill-marketplace` | Browse and install skills from the community catalog |
| `/index` | Index the current codebase for semantic search (via ZeroDB) |

## Other

| Command | Description |
|---------|-------------|
| `/security` | Scan project config files for vulnerabilities (score 0–100) |
| `/bridge` | Start a remote session via Agent Cloud |
| `/voice` | Speech-to-text input (Whisper STT) |
| `/plan` | Enter read-only planning mode |
| `/brief` | Toggle compact message layout |

## CLI Flags

These flags are passed when launching Cody from the terminal:

```bash
cody [flags] [-p "prompt"]
```

| Flag | Description |
|------|-------------|
| `--model <id>` | Use a specific model |
| `--fast` | Enable fast mode |
| `-p, --prompt <text>` | Single prompt (non-interactive) |
| `--print` | Print mode — pipe-friendly output |
| `--dry-run` | Preview what would be sent to the LLM without making an API call |
| `--color` | Force or disable color output |
| `--output-format markdown` | Structured markdown output (useful for CI/CD) |
| `--api-key <key>` | Pass API key inline |
| `--safe-mode` | Disable plugins and custom skills |

## Shell Escape

Run a shell command from within Cody and get an automatic explanation of the output:

```
! npm test
```

Triggers auto-explain on 4+ lines of output — explains errors, test failures, and build output.
