CLI Reference

Complete documentation for all YOLO Developer command-line interface commands.

Table of contents

  1. Overview
    1. Basic Usage
    2. Interactive Mode
    3. Global Options
  2. Command Summary
  3. yolo chat
    1. Synopsis
    2. Examples
  4. yolo integrate
    1. Synopsis
    2. Options
    3. Examples
  5. yolo init
    1. Synopsis
    2. Options
    3. Git Repository Features
    4. Examples
    5. Files Created
  6. yolo seed
    1. Synopsis
    2. Arguments
    3. Options
    4. Examples
    5. Input Formats
  7. yolo run
    1. Synopsis
    2. Options
    3. Examples
    4. Agent Execution Order
  8. yolo status
    1. Synopsis
    2. Options
    3. Examples
  9. yolo logs
    1. Synopsis
    2. Options
    3. Examples
  10. yolo config
    1. Synopsis
    2. Commands
    3. Examples
  11. yolo tools
    1. Synopsis
    2. Commands
    3. Options
    4. Examples
    5. Configuration
  12. yolo tune
    1. Synopsis
    2. Options
    3. Examples
  13. yolo mcp
    1. Synopsis
    2. Options
    3. Examples
  14. yolo scan
    1. Synopsis
    2. Options
    3. Examples
  15. yolo git
  16. yolo pr
  17. yolo issue
  18. yolo release
  19. yolo workflow
  20. yolo import
  21. yolo gather
  22. yolo web
  23. Exit Codes
  24. Exit Codes
  25. Environment Variables
  26. Next Steps

Overview

The YOLO Developer CLI (yolo) provides commands for initializing projects, seeding requirements, running autonomous development sprints, and monitoring progress.

Basic Usage

yolo [OPTIONS] COMMAND [ARGS]...

Interactive Mode

Run yolo with no arguments to start an interactive chat session:

yolo

Or explicitly:

yolo chat

For one-shot prompts:

yolo "Summarize the current sprint status"

Global Options

Option Description
--version Show version and exit
--help Show help message and exit
-v, --verbose Enable verbose output
-q, --quiet Suppress non-essential output
--config PATH Use specific config file

Command Summary

Command Description
yolo init Initialize a new YOLO project
yolo chat Start interactive chat or run one-shot prompts
yolo integrate Integrate MCP clients (Codex, Claude Code, Cursor, VS Code)
yolo seed Seed requirements for development
yolo run Execute autonomous development sprint
yolo status Display current sprint status
yolo logs View agent activity logs
yolo config Manage project configuration
yolo tools Manage external CLI tool integrations
yolo tune Adjust quality thresholds
yolo mcp Start MCP server for Claude Code
yolo scan Scan existing project for brownfield context
yolo git Local git operations
yolo pr Pull request operations
yolo issue Issue operations
yolo release Release operations
yolo workflow GitHub workflow automation
yolo import Import GitHub issues
yolo gather Interactive requirements gathering
yolo web Web dashboard

yolo chat

Start an interactive chat session or run a one-shot prompt.

Synopsis

yolo chat [PROMPT...]

Examples

Interactive mode:

yolo chat

One-shot prompt:

yolo chat "Summarize the current sprint status"

Pipe input:

echo "Draft release notes" | yolo chat

yolo integrate

Configure MCP client settings for external AI tools.

Synopsis

yolo integrate <client> [OPTIONS]

Options

Option Description
--config-path, -c Override default config path
--project-dir, -p Project directory for uv run fallback
--dry-run Show JSON without writing
--force Overwrite existing MCP entry
--yes, -y Skip confirmation prompt

Examples

yolo integrate claude-code
yolo integrate codex --dry-run
yolo integrate cursor --config-path /custom/settings.json --yes
yolo integrate vscode --force

yolo init

Initialize a new YOLO Developer project in the current directory.

Synopsis

yolo init [OPTIONS] [PATH]

Options

Option Type Default Description
--name, -n TEXT Directory name Project name
--author, -a TEXT git config Author name for pyproject.toml
--email, -e TEXT git config Author email for pyproject.toml
--interactive, -i FLAG False Prompt for project details
--no-input FLAG False Use defaults without prompting
--existing, --brownfield FLAG False Add YOLO to existing project
--scan-only FLAG False Scan existing project without changes
--non-interactive FLAG False Skip brownfield prompts
--hint TEXT None Hint about project type
--skip-git FLAG False Skip git repository prompts
--skip-github FLAG False Skip GitHub repository creation prompts

Git Repository Features

The init command automatically handles git repository setup:

  1. Detection: Checks if directory is a git repository
  2. Initialization: Offers to run git init if not initialized
  3. Remote Detection: Displays configured remote repositories
  4. GitHub Creation: Creates a new GitHub repository via gh CLI (if available)
  5. Initial Commit: Creates an initial commit with project files
  6. Push: Pushes initial commit to remote repository

Examples

Basic initialization:

yolo init

Output:

Initializing YOLO Developer project...

? Project name: my-api
? Author name: Developer
? Author email: dev@example.com

This directory is not a git repository. Initialize git? [Y/n] y
Git repository initialized!

           Git Repository Status
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┓
┃ Property     ┃ Value               ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━┩
│ Git Status   │ Initialized         │
│ Remotes      │ None configured     │
└──────────────┴─────────────────────┘

Remote Repository Options:
  1. Create a new GitHub repository
  2. Add an existing remote URL
  3. Skip remote setup
Choose an option [3]:

Creating configuration...
  ✓ Created yolo.yaml
  ✓ Created .yolo/ directory
  ✓ Initialized memory store

Create initial commit with project files? [Y/n] y
Initial commit created!

Project initialized successfully!

Skip git prompts:

yolo init --skip-git

Skip GitHub prompts only:

yolo init --skip-github

Fully automated (no prompts):

yolo init --no-input

Brownfield initialization:

yolo init --brownfield

Scan only (no changes):

yolo init --brownfield --scan-only

Files Created

File Description
yolo.yaml Project configuration
.yolo/ YOLO data directory
.yolo/memory/ ChromaDB vector store
.yolo/cache/ Agent response cache

yolo seed

Seed requirements for autonomous development.

Synopsis

yolo seed [OPTIONS] [FILE]

Arguments

Argument Type Required Description
FILE PATH No* Path to requirements document

*Required unless --text is provided.

Options

Option Type Default Description
--text, -t TEXT None Provide requirements as inline text
--format, -f CHOICE auto Input format: auto, markdown, text
--validate-only FLAG False Validate without storing seed
--skip-validation FLAG False Skip ambiguity/contradiction checks
--output, -o PATH None Export parsed requirements to file

Examples

Seed from file:

yolo seed requirements.md

Output:

Parsing requirements document...
  ✓ Parsed 12 requirements

Running validation...
  ✓ No ambiguities detected
  ✓ No contradictions found
  ✓ SOP constraints validated

Seed Summary:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  Requirements: 12 total
    - Functional: 8
    - Non-Functional: 3
    - Constraints: 1
  Quality Score: 0.94

Seed ID: seed_a1b2c3d4-e5f6-7890-abcd-ef1234567890

Ready to run: yolo run

Seed from inline text:

yolo seed --text "Build a REST API with user authentication using JWT tokens"

Validate without seeding:

yolo seed requirements.md --validate-only

Output with ambiguities detected:

Parsing requirements document...
  ✓ Parsed 8 requirements

Running validation...
  ⚠ 2 ambiguities detected:
    1. "fast response times" - What is considered fast? (< 100ms? < 500ms?)
    2. "support many users" - Specific concurrency target needed

  ✓ No contradictions found
  ✓ SOP constraints validated

Clarification questions generated:
  Q1: What response time threshold defines "fast"?
      Suggestions: < 100ms, < 200ms, < 500ms
  Q2: How many concurrent users should be supported?
      Suggestions: 100, 1000, 10000

Fix ambiguities and re-run, or use --skip-validation to proceed.

Input Formats

Markdown (recommended):

# Project Requirements

## Functional Requirements
- User can register with email
- User can login with password

## Non-Functional Requirements
- Response time < 200ms
- 99.9% uptime

## Constraints
- Must use PostgreSQL

Plain text:

User Management System

The system must allow users to register with their email address.
Users must be able to login with a password.
Response times should be under 200ms.
The system must use PostgreSQL for data storage.

yolo run

Execute an autonomous development sprint.

Synopsis

yolo run [OPTIONS]

Options

Option Type Default Description
--dry-run, -d FLAG False Validate configuration and seed without running
--verbose, -v FLAG False Show detailed output
--json, -j FLAG False Output JSON summary
--resume, -r FLAG False Resume from last checkpoint
--thread-id, -t TEXT None Use specific checkpoint thread ID
--continue FLAG False Continue from last checkpoint (alias for --resume)
--agents TEXT None Comma-separated list of agents to run (accepted but not yet enforced)
--max-iterations INT None Maximum iterations per agent (accepted but not yet enforced)
--timeout INT None Timeout in seconds per agent (accepted but not yet enforced)
--watch FLAG False Watch mode (accepted but not yet enforced)
--output-dir PATH None Output directory for artifacts (accepted but not yet enforced)

Examples

Dry run (no changes):

yolo run --dry-run

Resume from checkpoint:

yolo run --resume

Use a specific thread ID:

yolo run --thread-id my-session

Agent Execution Order

  1. Analyst - Requirement crystallization
  2. PM - Story generation
  3. Architect - System design
  4. Dev - Implementation
  5. TEA - Test validation
  6. SM - Orchestration (runs throughout)

yolo status

Display current sprint status and progress.

Synopsis

yolo status [OPTIONS]

Options

Option Type Default Description
--format, -f CHOICE table Output format: table, json, yaml
--json, -j FLAG False Output JSON (alias for --format json)
--health, -H FLAG False Show only health metrics
--sessions, -s FLAG False Show only session list
--verbose, -v FLAG False Show detailed health metrics

Note: --agents, --gates, --stories, --watch, and --refresh are accepted but currently ignored (a warning is shown).

Examples

Basic status:

yolo status

JSON output for scripting:

yolo status --format json

yolo logs

View agent activity logs and audit trail.

Synopsis

yolo logs [OPTIONS]

Options

Option Type Default Description
--agent, -a TEXT All Filter by agent name
--since, -s TEXT None Show logs since time (e.g., 1h, 30m)
--type, -t TEXT None Filter by decision type
--limit, -n INT 20 Maximum number of entries
--all, -A FLAG False Show all entries
--verbose, -v FLAG False Show full details
--json, -j FLAG False Output JSON to stdout
--format CHOICE text Output format: text or json
--export PATH None Export JSON output to a file

Examples

View recent logs:

yolo logs

Output:

2024-01-15 10:23:45 [INFO] [ANALYST] Starting requirement analysis
2024-01-15 10:23:47 [INFO] [ANALYST] Parsing 12 requirements
2024-01-15 10:24:12 [INFO] [ANALYST] Crystallized: "User registration" → 3 specs
2024-01-15 10:24:35 [INFO] [ANALYST] Crystallized: "User authentication" → 2 specs
2024-01-15 10:25:02 [WARN] [ANALYST] Ambiguity detected: "fast response"
2024-01-15 10:25:03 [INFO] [ANALYST] Generated clarification question
2024-01-15 10:26:19 [INFO] [ANALYST] Analysis complete: 12 requirements → 18 specs
2024-01-15 10:26:20 [INFO] [PM] Starting story generation
2024-01-15 10:26:45 [INFO] [PM] Created story: US-001 User Registration
2024-01-15 10:27:02 [INFO] [PM] Created story: US-002 User Authentication

Filter by agent:

yolo logs --agent dev

Export to JSON:

yolo logs --export audit.json --format json

Output (audit.json):

{
  "entries": [
    {
      "timestamp": "2024-01-15T10:23:45Z",
      "level": "INFO",
      "agent": "ANALYST",
      "message": "Starting requirement analysis",
      "context": {
        "seed_id": "seed_abc123",
        "requirement_count": 12
      }
    },
    {
      "timestamp": "2024-01-15T10:24:12Z",
      "level": "INFO",
      "agent": "ANALYST",
      "message": "Crystallized requirement",
      "context": {
        "original": "User registration",
        "specs_generated": 3,
        "confidence": 0.92
      }
    }
  ]
}

yolo config

Manage project configuration.

Synopsis

yolo config [COMMAND] [OPTIONS]

Commands

Command Description
show Display current configuration
set Set a configuration value
get Get a specific configuration value
reset Reset to default configuration
validate Validate configuration file
export Export configuration to a YAML file
import Import configuration from a YAML file

Examples

Show configuration:

yolo config show

Output:

project_name: my-api
llm:
  provider: auto
  cheap_model: gpt-5.2-instant
  premium_model: claude-sonnet-4-20250514
  best_model: claude-opus-4-5-20251101
  openai_api_key: "**********" (configured)
  anthropic_api_key: "**********" (configured)
  openai:
    code_model: gpt-5.2-pro
  hybrid:
    enabled: false
quality:
  test_coverage_threshold: 0.8
  gate_pass_threshold: 0.7
memory:
  persist_path: .yolo/memory
  vector_store_type: chromadb
  graph_store_type: json
agents:
  max_iterations: 10
  timeout_seconds: 300

Set a value:

yolo config set quality.test_coverage_threshold 0.9

Get a specific value:

yolo config get llm.premium_model

Output:

claude-sonnet-4-20250514

Validate configuration:

yolo config validate

Output:

Validating yolo.yaml...
  ✓ Schema valid
  ✓ API keys configured
  ✓ Memory directory writable
  ✓ All settings valid

Configuration is valid.

Export configuration:

yolo config export -o yolo-config-export.yaml

Import configuration:

yolo config import yolo-config-export.yaml

yolo tools

Manage external CLI tool integrations like Claude Code and Aider.

Synopsis

yolo tools [OPTIONS] [COMMAND]

Commands

Command Description
status Show tool availability and configuration

Options

Option Type Default Description
--json, -j FLAG False Output as JSON

Examples

Show tool status:

yolo tools

Output:

                    External CLI Tools
┏━━━━━━━━━━━━━━┳━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━┓
┃ Tool         ┃ Enabled ┃ Available ┃ Binary              ┃ Timeout ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━┩
│ claude_code  │   Yes   │    Yes    │ /usr/local/bin/claude │  300s │
│ aider        │   No    │    No     │ aider (not found)     │  300s │
└──────────────┴─────────┴───────────┴─────────────────────┴─────────┘

JSON output:

yolo tools --json

Output:

{
  "tools": [
    {
      "name": "claude_code",
      "enabled": true,
      "binary": "claude",
      "available": true,
      "path": "/usr/local/bin/claude",
      "timeout": 300,
      "output_format": "json"
    },
    {
      "name": "aider",
      "enabled": false,
      "binary": "aider",
      "available": false,
      "path": null,
      "timeout": 300,
      "output_format": "json"
    }
  ],
  "status": "success"
}

Configuration

Tools are configured in yolo.yaml:

tools:
  claude_code:
    enabled: true
    path: /custom/path/to/claude  # optional
    timeout: 300
    output_format: json
    extra_args: []
  aider:
    enabled: false

Or via environment variables:

export YOLO_TOOLS__CLAUDE_CODE__ENABLED=true
export YOLO_TOOLS__CLAUDE_CODE__TIMEOUT=600
export YOLO_TOOLS__AIDER__ENABLED=true

yolo tune

Customize agent templates and behaviors.

Synopsis

yolo tune [AGENT_NAME] [OPTIONS]

Options

Option Type Default Description
--list, -l FLAG False List all configurable agents
--edit, -e FLAG False Edit the agent template in $EDITOR
--reset, -r FLAG False Reset the agent template to defaults
--export PATH None Export the agent template to a file
--import PATH None Import the agent template from a file
--json, -j FLAG False Output template as JSON

Note: quality threshold tuning is not yet supported in this command. Use yolo config set quality.* to adjust thresholds.

Examples

List agents:

yolo tune --list

Show an agent template:

yolo tune analyst

Edit an agent template:

yolo tune analyst --edit

Export/import templates:

yolo tune analyst --export analyst.yaml
yolo tune analyst --import analyst.yaml

yolo mcp

Start MCP server for Claude Code integration.

Synopsis

yolo mcp [OPTIONS]

Options

Option Type Default Description
--transport, -t CHOICE stdio Transport: stdio, http
--port, -p INT 8080 Port for HTTP transport
--host TEXT 127.0.0.1 Host for HTTP transport

Examples

Start with STDIO (for Claude Desktop):

yolo mcp

Start with HTTP transport:

yolo mcp --transport http --port 8080

Output:

Starting YOLO Developer MCP Server...
  Transport: HTTP
  Address: http://127.0.0.1:8080

Available tools:
  - yolo_seed: Provide seed requirements
  - yolo_run: Execute sprint
  - yolo_status: Get status
  - yolo_audit: Access audit

Server ready. Press Ctrl+C to stop.

See MCP Integration for detailed usage.


yolo scan

Scan an existing repository and optionally write .yolo/project-context.yaml.

Synopsis

yolo scan [OPTIONS] [PATH]

Options

Option Type Default Description
--scan-depth INT 3 Directory depth to scan
--max-files INT 1000 Maximum files to analyze
--git-history/--no-git-history FLAG from config Include git history analysis
--interactive, -i FLAG False Prompt for ambiguous findings
--hint TEXT None Hint about project type
--refresh FLAG False Overwrite existing project-context.yaml
--write-context/--no-write-context FLAG True Write project context file

Examples

yolo scan
yolo scan --refresh
yolo scan --max-files 200 --scan-depth 4

yolo git

Manage local Git operations.

yolo git status
yolo git commit -m "feat: update"
yolo git push

yolo pr

Manage pull requests.

yolo pr create --title "Title" --body "Body"
yolo pr merge 123 --method squash

yolo issue

Manage GitHub issues.

yolo issue create --title "Bug" --body "Details"
yolo issue close 123 --comment "Fixed in PR #456"

yolo release

Create GitHub releases.

yolo release create --tag v1.2.0 --name "Release 1.2.0" --body "Notes"

yolo workflow

Automate story workflows.

yolo workflow start US-001 --title "Add endpoint" --description "..."
yolo workflow complete US-001 --title "Add endpoint" --description "..." --commit "feat: add endpoint"

yolo import

Import GitHub issues and convert them into user stories.

yolo import issue 42
yolo import preview 42
yolo import issues --label "ready" --auto-seed

Common options:

  • --repo owner/repo: override repository
  • --auto-seed: write seed file to .yolo/imported-issues/
  • --preview: preview only
  • --output: export story to file

yolo gather

Interactive requirements gathering with the Analyst agent.

yolo gather start my-project --description "Build a task manager"
yolo gather list
yolo gather export 20250122093000 --format markdown --output requirements.md

yolo web

Start the local web dashboard and API.

yolo web start
yolo web start --port 8080 --host 0.0.0.0

Exit Codes

Exit Codes

Code Description
0 Success
1 General error
2 Configuration error
3 Validation error
4 API error (LLM)
5 Quality gate failure
130 Interrupted (Ctrl+C)

Environment Variables

All configuration can be overridden via environment variables using the YOLO_ prefix:

Variable Description
YOLO_PROJECT_NAME Project name
YOLO_LLM__PROVIDER Primary LLM provider (auto/openai/anthropic/hybrid)
YOLO_LLM__CHEAP_MODEL Model for routine tasks
YOLO_LLM__PREMIUM_MODEL Model for complex tasks
YOLO_LLM__BEST_MODEL Model for critical tasks
YOLO_LLM__OPENAI__API_KEY OpenAI API key (preferred)
YOLO_LLM__OPENAI_API_KEY OpenAI API key (legacy)
YOLO_LLM__ANTHROPIC_API_KEY Anthropic API key
YOLO_LLM__OPENAI__CODE_MODEL OpenAI model for code tasks
YOLO_LLM__HYBRID__ENABLED Enable hybrid routing
YOLO_QUALITY__TEST_COVERAGE_THRESHOLD Coverage threshold
YOLO_QUALITY__GATE_PASS_THRESHOLD Gate pass threshold
YOLO_MEMORY__PERSIST_PATH Memory storage path
YOLO_MEMORY__VECTOR_STORE_TYPE Vector store backend
YOLO_MEMORY__GRAPH_STORE_TYPE Graph store backend
YOLO_BROWNFIELD__SCAN_DEPTH Brownfield scan depth
YOLO_BROWNFIELD__MAX_FILES_TO_ANALYZE Brownfield scan file limit
YOLO_GITHUB__TOKEN GitHub token
YOLO_GITHUB__REPOSITORY GitHub repo slug (owner/repo)

Next Steps


Table of contents


Back to top

Copyright © 2024-2026 YOLO Developer. Distributed under the MIT License.