Skip to main content
AI/MLathola

setup

Provisions the oracle ML inference daemon with onnxruntime via uv. Use when setting up local ONNX model inference for skill quality evaluation.

Stars
294
Source
athola/claude-night-market
Updated
2026-05-30
Slug
athola--claude-night-market--setup
View on GitHubRaw SKILL.md

// install — copy + paste into any project

mkdir -p .claude/skills && curl -fsSL https://raw.githubusercontent.com/athola/claude-night-market/HEAD/plugins/oracle/skills/setup/SKILL.md -o .claude/skills/setup.md

Drops the SKILL.md into .claude/skills/setup.md. Works with Claude Code, Cursor, and any agent that loads SKILL.md files from .claude/skills/.

Oracle Setup

Provision the ML inference environment.

When NOT To Use

  • The daemon is already provisioned and only needs to be used
  • Evaluating skill quality itself (use abstract:skills-eval)

What This Does

  1. Creates a Python 3.11+ virtual environment using uv
  2. Installs onnxruntime into the venv
  3. Verifies the installation

Prerequisites

  • uv must be installed
  • Internet connection for initial download

Steps

  1. Run provisioning:
cd plugins/oracle && uv run python -c "
from oracle.provision import provision_venv, get_venv_path
result = provision_venv(get_venv_path())
print(result.message)
"
  1. Report result to the user.
  2. If successful, tell the user the daemon will start on next session.
  3. If failed, show the error and suggest checking uv and network.

Exit Criteria

  • provision_venv() returns a result with a non-error message and the venv path exists on disk under plugins/oracle/
  • onnxruntime is importable inside the provisioned venv; verified by uv run python -c "import onnxruntime" exiting 0
  • User receives a clear success message stating the daemon will start on next session, or a clear failure message citing the specific error and whether it is a uv or network issue
  • On failure, no partial venv left in a broken state that would prevent a clean retry