Initial commit: llm-router — smart OpenAI/Anthropic/Ollama router

A stdlib pre-router in front of Ollama with LiteLLM backend:
- auto model selection by content/tools/modality, with fallbacks
- OpenAI /v1, Anthropic /v1/messages, and Ollama-native /api/* endpoints
- Whisper-shaped /v1/audio/transcriptions + in-chat audio
- key-based fleet policies (e.g. force a client onto uncensored models)
- optional Bearer auth; launchd/systemd service install
- benchmark harnesses (speed, quality, agentic tool use) with sample results

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

scripts/setup.sh

@@ -0,0 +1,23 @@
+#!/usr/bin/env bash
+# One-time setup: create the Python 3.12 venv and install LiteLLM + Pillow.
+# The pre-router itself needs no venv (stdlib only); this is for the backend proxy.
+set -euo pipefail
+cd "$(dirname "$0")/.."
+
+PY="${PYTHON:-python3.12}"
+if ! command -v "$PY" >/dev/null 2>&1; then
+  # common macOS Homebrew fallback
+  [ -x /opt/homebrew/opt/python@3.12/bin/python3.12 ] && PY=/opt/homebrew/opt/python@3.12/bin/python3.12
+fi
+command -v "$PY" >/dev/null 2>&1 || { echo "Need Python 3.12. Set \$PYTHON, or:"; \
+  echo "  macOS:  brew install python@3.12"; echo "  Linux:  install python3.12"; exit 1; }
+
+echo ">> using $($PY --version) at $(command -v "$PY" 2>/dev/null || echo "$PY")"
+[ -d .venv ] || "$PY" -m venv .venv
+.venv/bin/pip install -q --upgrade pip
+.venv/bin/pip install -q -r requirements.txt
+.venv/bin/python -c "import litellm, PIL" && echo ">> deps OK (litellm, pillow)"
+echo ""
+echo "Next:"
+echo "  ./scripts/pull-models.sh   # pull a model fleet (edit the list first)"
+echo "  ./run.sh                   # start LiteLLM + router"