Prerequisites

Before installing Fortified LLM Client, ensure your system meets these requirements.

Required

Rust 1.70 or Later

Check if installed:

1
2

rustc --version
# Should show 1.70.0 or higher

Install Rust:

1
2

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source $HOME/.cargo/env

Visit rustup.rs for alternative installation methods.

Rust Nightly (for development/formatting)

The project uses nightly rustfmt for code formatting:

1

rustup toolchain install nightly

This is only required if you plan to contribute code or run cargo +nightly fmt.

Optional

Docling CLI (for PDF extraction)

Required only if you plan to use PDF extraction features (--pdf-input flag).

Install via pip:

1

pip install docling

Verify installation:

1

docling --version

Without Docling: You can still use all other features (LLM invocation, guardrails, JSON schema validation, etc.). PDF extraction will fail with an error if Docling is not installed.

LLM Provider Access

You need access to at least one LLM provider:

Option 1: OpenAI API

  • Sign up at platform.openai.com
  • Create an API key
  • Set environment variable: export OPENAI_API_KEY=sk-...

Option 2: Ollama (Local Models)

  • Install Ollama from ollama.com
  • Pull a model: ollama pull llama3
  • Start Ollama server: ollama serve (or run as background service)
  • Default API URL: http://localhost:11434/v1/chat/completions

Option 3: Other OpenAI-Compatible APIs

Any API compatible with OpenAI’s /v1/chat/completions endpoint works:

  • Azure OpenAI
  • Together AI
  • Anyscale
  • Custom hosted models

System Requirements

  • OS: Linux, macOS, or Windows (WSL recommended for Windows)
  • Memory: 512MB minimum (depends on LLM provider)
  • Disk: ~200MB for compiled binary
  • Network: Internet access for API-based providers (not needed for local Ollama)

Next Steps

Once your system meets these requirements, proceed to Installation.