2538 Chat 6cc837b6

# Chat

Interactive terminal chat that lets you converse with an LLM about your codebase while it invokes skene-growth tools to gather information.

## Prerequisites

- An API key configured (see [Configuration](configuration.md)) or a local LLM running
- A codebase to analyze

## Basic usage

```bash
# Chat about the current directory
uvx skene-growth chat

# Chat about a specific codebase
uvx skene-growth chat /path/to/project

# Using the `skene` shorthand (defaults to chat)
uvx --from skene-growth skene
```

The `skene` entry point defaults to the `chat` command, providing a convenient shorthand for interactive sessions.

## Flags

| Flag | Short | Description | Default |
|------|-------|-------------|---------|
| `--api-key` | | API key for LLM provider | `SKENE_API_KEY` env var |
| `--provider` | `-p` | LLM provider | Config or `openai` |
| `--model` | `-m` | LLM model name | Provider default |
| `--base-url` | | Base URL for OpenAI-compatible API endpoint. Required when provider is `generic`. | `SKENE_BASE_URL` env var |
| `--max-steps` | | Maximum tool calls per user request | `4` |
| `--tool-output-limit` | | Max tool output characters kept in context | `4000` |
| `--debug` | | Log all LLM input/output to `.skene-growth/debug/` | Off |

## How it works

The chat command starts an interactive terminal session where:

1. You type a question or request about your codebase
2. The LLM decides which skene-growth tools to call (analyze, search, read files, etc.)
3. Tool results are fed back to the LLM within the context window
4. The LLM synthesizes a response based on the tool outputs

The `--max-steps` flag controls how many tool calls the LLM can make per request. Increase this for complex queries that require multiple analysis passes. The `--tool-output-limit` flag controls how much of each tool's output is kept in context to avoid exceeding token limits.

## Tips for effective use

- **Be specific** — "What growth features does this codebase have?" works better than "Tell me about this code"
- **Increase max-steps for deep analysis** — Use `--max-steps 8` when you want the LLM to do thorough multi-step analysis
- **Use debug mode to understand behavior** — `--debug` logs all LLM interactions so you can see what tools are being called

## Next steps

- [Analyze command](analyze.md) — Run a full codebase analysis
- [LLM providers](llm-providers.md) — Configure different providers for chat
- [Configuration](configuration.md) — Set defaults so you don't need flags every time