CLI Reference

Complete documentation for all AIGovHub CLI commands.

Global Options

These options are available for all commands:

aigovhub [OPTIONS] COMMAND [ARGS]

Option	Description
`--version`, `-v`	Show version and exit
`--help`	Show help message and exit
`--install-completion`	Install shell completion
`--show-completion`	Show completion script

Commands

`aigovhub scan`

Scan a repository to detect AI systems and generate an AI-SBoM artifact.

aigovhub scan [PATH] [OPTIONS]

Arguments

Argument	Description	Default
`PATH`	Path to repository to scan	`.` (current directory)

Options

Option	Short	Description	Default
`--output`	`-o`	Output file path	`./aigovhub.yaml`
`--format`	`-f`	Output format (`yaml` or `json`)	`yaml`
`--confidence`	`-c`	Minimum confidence threshold (0.0-1.0)	`0.7`
`--llm/--no-llm`		Enable/disable LLM fallback	`--llm`
`--llm-provider`		LLM provider (`anthropic`, `openai`, `local`)	From config
`--dry-run`	`-n`	Preview without writing output	`false`
`--verbose`		Show detailed output	`false`

Examples

# Basic scan of current directory
aigovhub scan .
 
# Scan with custom output location
aigovhub scan /path/to/repo --output reports/compliance.yaml
 
# Scan with higher confidence threshold
aigovhub scan . --confidence 0.9
 
# Fast scan without LLM
aigovhub scan . --no-llm
 
# Preview what would be detected
aigovhub scan . --dry-run --verbose
 
# JSON output for processing
aigovhub scan . --format json --output ai-systems.json
 
# Use specific LLM provider
aigovhub scan . --llm-provider openai

Exit Codes

Code	Meaning
0	Success
1	Error during scanning

`aigovhub init`

Initialize a new aigovhub.yaml configuration file.

aigovhub init [PATH] [OPTIONS]

Arguments

Argument	Description	Default
`PATH`	Path to initialize	`.` (current directory)

Options

Option	Short	Description	Default
`--force`	`-f`	Overwrite existing file	`false`
`--template`	`-t`	Template to use	`standard`

Templates

Template	Description
`minimal`	Bare minimum fields only
`standard`	Common fields with scan config
`enterprise`	Full schema with example AI system

Examples

# Initialize with standard template
aigovhub init
 
# Initialize in specific directory
aigovhub init /path/to/repo
 
# Force overwrite existing file
aigovhub init --force
 
# Use minimal template
aigovhub init --template minimal
 
# Use enterprise template with examples
aigovhub init --template enterprise

Exit Codes

Code	Meaning
0	Success
1	File exists (use `--force`) or other error

`aigovhub validate`

Validate an aigovhub.yaml file against the schema.

aigovhub validate [PATH] [OPTIONS]

Arguments

Argument	Description	Default
`PATH`	Path to file or directory	`.` (looks for `aigovhub.yaml`)

Options

Option	Short	Description	Default
`--strict`	`-s`	Fail on warnings	`false`
`--schema-version`		Validate against specific version	Latest

Examples

# Validate aigovhub.yaml in current directory
aigovhub validate
 
# Validate specific file
aigovhub validate reports/compliance.yaml
 
# Strict validation (warnings become errors)
aigovhub validate --strict
 
# Validate against specific schema version
aigovhub validate --schema-version 1.0.0

Exit Codes

Code	Meaning
0	Valid
1	Invalid or errors found

Validation Checks

The validator checks:

Schema compliance - Required fields, correct types
Semantic validity - Valid risk categories, confidence ranges
Consistency - No duplicate AI system IDs
Warnings - Empty AI systems list, unknown fields

`aigovhub benchmark`

Run benchmark evaluation against a labeled dataset.

Note: This command is not yet implemented and will display a placeholder message. It is planned for a future release.

aigovhub benchmark [OPTIONS]

Options

Option	Short	Description	Default
`--dataset`	`-d`	Path to benchmark dataset YAML	`benchmark/repos.yaml`
`--output`	`-o`	Directory for results	`benchmark/results/`
`--no-cache`		Skip cached repository data	`false`

Planned Examples (Not Yet Working)

# Run benchmark with default dataset
aigovhub benchmark
 
# Use custom dataset
aigovhub benchmark --dataset my-dataset.yaml
 
# Force fresh clone of repositories
aigovhub benchmark --no-cache

Shell Completion

AIGovHub CLI supports shell completion for bash, zsh, and fish.

Install Completion

# Detect shell and install
aigovhub --install-completion
 
# Or manually for specific shells
aigovhub --show-completion bash >> ~/.bashrc
aigovhub --show-completion zsh >> ~/.zshrc
aigovhub --show-completion fish >> ~/.config/fish/completions/aigovhub.fish

Using Completion

After installation, restart your shell or source your config:

source ~/.bashrc  # or ~/.zshrc

Now you can use tab completion:

aigovhub sc<TAB>        # completes to "scan"
aigovhub scan --<TAB>   # shows available options

Output Formats

YAML (Default)

Human-readable, suitable for version control:

schema_version: "1.0.0"
generated_at: "2025-01-15T10:30:00Z"
ai_systems:
  - id: "ai-001"
    name: "model"
    type: "ml_model"

JSON

Machine-readable, suitable for processing:

{
  "schema_version": "1.0.0",
  "generated_at": "2025-01-15T10:30:00Z",
  "ai_systems": [
    {
      "id": "ai-001",
      "name": "model",
      "type": "ml_model"
    }
  ]
}

Environment Variables

The CLI respects these environment variables:

Variable	Description	Default
`AIGOVHUB_CONFIDENCE_THRESHOLD`	Default confidence threshold	`0.7`
`AIGOVHUB_USE_LLM_FALLBACK`	Enable LLM by default	`true`
`AIGOVHUB_LLM_PROVIDER`	Default LLM provider	`anthropic`
`AIGOVHUB_DEFAULT_OUTPUT_FILE`	Default output filename	`aigovhub.yaml`
`ANTHROPIC_API_KEY`	Anthropic API key	-
`OPENAI_API_KEY`	OpenAI API key	-

See Configuration for detailed configuration options.