scan

Analyzes repository statistics including file counts, language distribution, token counting, and optional security scanning.

Usage

infiniloom scan [PATH] [OPTIONS]

Description

The scan command traverses a repository respecting .gitignore, detects programming languages, counts tokens using tiktoken, and optionally scans for secrets and API keys.

Options

--model <MODEL>

Target model for token counting (default: claude). Supports claude, gpt4, gpt4o, gemini, llama, mistral, deepseek, qwen, cohere, grok

--hidden

Include hidden files in scan

--verbose

Show detailed file list

--json

Output results as JSON

--security-check

Scan for secrets and API keys

--sample <N>

Sample N random files for estimation (useful for large repos)

--sample-percent <P>

Sample P% of files for estimation

-i, --include <PATTERN>

Include only files matching glob pattern (repeatable)

-e, --exclude <PATTERN>

Exclude files/directories matching pattern (repeatable)

--include-tests

Include test files in scan (normally excluded)

Examples

# Quick repository scan
infiniloom scan

# Scan with security check
infiniloom scan --security-check

# JSON output for CI/CD
infiniloom scan --json

# Verbose scan with file list
infiniloom scan --verbose

Output

File counts by language
Total token count
Language distribution
Security findings (if enabled)