Skip to main content
Version: Next (unreleased)

Getting Started

Installation​

GrillMyCode is a GitHub Action — there is nothing to install. Add a workflow file to the student or assessment repository to get started.

tip

Not sure which inputs to use? The Workflow Wizard walks you through each option and generates the YAML for you.

Minimal workflow​

name: GrillMyCode

on:
  push:
    branches: ["main", "master"]
  workflow_dispatch:

# A new push cancels any run still in progress for the same branch,
# so only the latest commit is ever assessed (see FAQ).
# Do not modify this setting unless you have a compelling reason to.
concurrency:
  group: grillmycode-${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true

jobs:
  generate-questions:
    runs-on: ubuntu-latest
    timeout-minutes: 15
    permissions:
      contents: write  # gmc-assessments release + PDF asset
      issues: write    # assessment issue
    steps:
      - uses: actions/checkout@v6
        with:
          fetch-depth: 0    # full history required for diff resolution

      - uses: NSCC-ITC-Assessment/GrillMyCode@v1
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}
          api_key: ${{ secrets.OPENROUTER_API_KEY }}
          # If desired, uncomment this input and edit to use a different one —
          # any model from https://openrouter.ai/models (provider/model-name).
          # ai_model: "google/gemini-3.5-flash-lite"

Copy this file to .github/workflows/grill-my-code.yml in the student repository.

This workflow needs one secret: OPENROUTER_API_KEY. Question generation runs through OpenRouter, which requires its own API key — the built-in GITHUB_TOKEN cannot be used for it. Add the key once as an organisation-level Actions secret and every student repository inherits it automatically. See the OpenRouter setup guide for the one-time steps.

Trigger event​

The push trigger fires whenever a commit lands on main or master — whether pushed directly or merged in via a pull request. Both paths are treated identically; the assessed diff is always the student's full work history on the default branch.

workflow_dispatch: allows a manual re-run from the Actions tab without pushing a new commit.

A manual run uses the settings baked into the workflow file. To change a setting for a single run — a different question count or model, a retargeted instructor context — expose it as a workflow_dispatch input and it becomes a form field on the Run workflow button. See Manual Run Overrides.

See Example Workflows for ready-to-use files.

Customising the questions​

Use instructor_context to give the AI assignment-specific instructions:

- uses: NSCC-ITC-Assessment/GrillMyCode@v1
  with:
    github_token: ${{ secrets.GITHUB_TOKEN }}
    api_key: ${{ secrets.OPENROUTER_API_KEY }}
    num_questions: '8'
    instructor_context: |
      Assignment 3 — Python loops. Prioritize execution flow questions
      that trace what a loop produces for a given input, conceptual
      questions about loop design choices, and at least one error
      identification question about off-by-one errors.

Choosing a model​

OpenRouter is the only supported provider, so ai_provider can be left out entirely. What you can choose is the model: ai_model accepts any OpenRouter model in provider/model-name format, and defaults to google/gemini-3.5-flash-lite. See recommended models for tested, low-cost options.

Next steps​