# Raw documentation

> Raw is a local coding agent for models with limited context. Learn how to install it, configure models and agents, and run it from the terminal, a local dashboard or ACP.

Raw (npm package `@tlelabs/raw`) is an open-source local coding agent from TLE Labs, released under the MIT license. It is built for models with limited context. You decide what each run can use: an agent selects an ordered list of tools, skills and hooks, a system prompt, and one model connection.

## What Raw does

- **Explicit capabilities.** Tools, skills, hooks and MCP tools run only when an agent selects them. Nothing is added implicitly.
- **Local and hosted models.** Connect a local model server such as Ollama, or a hosted API through one of four wire methods: OpenAI-compatible Chat Completions, OpenAI Responses, Anthropic Messages, or Google Generate Content.
- **Saved sessions.** Conversations persist across process restarts and can be resumed from any surface.
- **Extensible parts.** Write your own tools, skills, hooks and agents, and package them for another machine.

Raw runs tools with your operating system account’s permissions. The working directory sets where relative paths resolve, but it is not a sandbox. Read [Permissions](https://raw.tlelabs.com/docs/guides/permissions/) before you run Raw on code you do not trust.

## Ways to use Raw

Raw has three surfaces that share the same agents and saved sessions:

- **Terminal.** Run one task with `raw "task"`, or open a saved REPL with `raw`.
- **Dashboard.** Run `raw dashboard` to chat in a local browser app, review approvals and context usage, and edit agents, tools, models and packages.
- **ACP.** Run `raw --acp --stdio` so an IDE or a parent agent can drive sessions over the Agent Client Protocol.

## Where to go next

[Install Raw](https://raw.tlelabs.com/docs/start/installation/): Requirements, install options and first-run configuration.

[Quickstart](https://raw.tlelabs.com/docs/start/quickstart/): Configure a model, run a task, open the REPL and the dashboard.

[Core concepts](https://raw.tlelabs.com/docs/start/concepts/): Models, agents, tools, skills, hooks, MCP, vars, packages and sessions.

[Models and agents](https://raw.tlelabs.com/docs/guides/models-and-agents/): Configure providers, methods, request controls and agent settings.

[Permissions](https://raw.tlelabs.com/docs/guides/permissions/): Understand OS permissions, tool rules and approvals.

[Dashboard](https://raw.tlelabs.com/docs/guides/dashboard/): Use the local browser app for chat, setup and packages.

[Terminal](https://raw.tlelabs.com/docs/guides/terminal/): Use the REPL, output options and the footer.

[Sessions](https://raw.tlelabs.com/docs/guides/sessions/): Resume conversations and manage saved history and retention.

[Context and compaction](https://raw.tlelabs.com/docs/guides/context-and-compaction/): Understand context estimates, compaction and cache behavior.

[Extend Raw](https://raw.tlelabs.com/docs/extend/tools/): Write tools, skills, hooks, MCP servers, variables, packages and ACP clients.

[Reference](https://raw.tlelabs.com/docs/reference/cli/): Command-line flags, exit codes and the configuration schema.