Commands & Flags

Usage

sylva [repo] [options]

The [repo] positional argument accepts an absolute path to a local repository. If omitted, Sylva prompts interactively or falls back to environment variables.

Options

Flag	Alias	Type	Default	Description
`--github-repository <url>`		String		Public GitHub repository URL to analyze
`--local-repository <path>`		String		Absolute path to a local repository
`-m, --model <model>`		String	Provider default	LLM model in `provider/model` format
`-i, --max-iterations <n>`		Number	`35`	Maximum RLM iterations for codebase analysis
`--list-models`		Boolean	`false`	List all supported models and exit
`--version`	`-V`	Boolean		Print version number
`--help`	`-h`	Boolean		Display help text

Examples

Analyze a Local Repository

# Explicit flag
sylva --local-repository /Users/me/projects/my-app -m openai/gpt-4o -i 5

# Positional argument
sylva /Users/me/projects/my-app

# Current directory (interactive prompt)
cd /Users/me/projects/my-app
sylva

Analyze a GitHub Repository

sylva --github-repository https://github.com/pallets/flask -m openai/gpt-4o -i 5

Sylva will:

Shallow-clone the repo into a temporary directory
Run the analysis pipeline
Save AGENTS.md to projects/<repo-name>/AGENTS.md
Delete the temporary clone

Complex Monorepo Analysis

sylva --local-repository . -m openai/gpt-5.2 -i 25

List Available Models

sylva --list-models

Output:

Supported models:
  GEMINI
    gemini/gemini-3.1-pro  (default)
    gemini/gemini-3.1-flash  (default mini)
    gemini/gemini-3-deep-think
  ANTHROPIC
    anthropic/claude-opus-4.6
    anthropic/claude-sonnet-4.6  (default)
    ...
  OPENAI
    openai/gpt-4o  (default)
    openai/gpt-4o-mini  (default mini)
    ...

Output Location

Generated files are saved to:

<cwd>/projects/<repo-name>/AGENTS.md

For example, if you analyze a project called my-app:

./projects/my-app/AGENTS.md

results matching ""

No results matching ""