diff --git a/.dockerignore b/.dockerignore
index 3eb8fac..a46f29f 100644
--- a/.dockerignore
+++ b/.dockerignore
@@ -7,4 +7,6 @@ Dockerfile
 bin/run-in-docker.sh
 bin/run-tests-in-docker.sh
 bin/run-tests.sh
+bin/validate-track-in-docker.sh
 tests/
+track/
diff --git a/.gitignore b/.gitignore
index c6f4076..90b351a 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,2 @@
 /tests/**/*/results.json
+track/
diff --git a/Dockerfile b/Dockerfile
index bd3578c..7213825 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,8 +1,51 @@
-FROM alpine:3.18
+FROM ubuntu:24.04 AS builder
 
-# install packages required to run the tests
-RUN apk add --no-cache jq coreutils
+ENV LUA_VER="5.4.8"
+ENV LUA_CHECKSUM="4f18ddae154e793e46eeab727c59ef1c0c0c2b744e7b94219710d76f530629ae"
+ENV LUAROCKS_VER="3.12.0"
+ENV LUAROCKS_GPG_KEY="3FD8F43C2BB3C478"
 
+RUN apt-get update && \
+    apt-get install -y curl gcc make unzip gnupg git && \
+    rm -rf /var/lib/apt/lists/* && \
+    apt-get purge --auto-remove && \
+    apt-get clean
+
+RUN curl -R -O -L http://www.lua.org/ftp/lua-${LUA_VER}.tar.gz && \
+    [ "$(sha256sum lua-${LUA_VER}.tar.gz | cut -d' ' -f1)" = "${LUA_CHECKSUM}" ] && \
+    tar -zxf lua-${LUA_VER}.tar.gz && \
+    cd lua-${LUA_VER} && \
+    make all install && \
+    cd .. && \
+    rm lua-${LUA_VER}.tar.gz && \
+    rm -rf lua-${LUA_VER}
+
+RUN curl -R -O -L https://luarocks.org/releases/luarocks-${LUAROCKS_VER}.tar.gz && \
+    curl -R -O -L https://luarocks.org/releases/luarocks-${LUAROCKS_VER}.tar.gz.asc && \
+    gpg --keyserver keyserver.ubuntu.com --recv-keys ${LUAROCKS_GPG_KEY} && \
+    gpg --verify luarocks-${LUAROCKS_VER}.tar.gz.asc luarocks-${LUAROCKS_VER}.tar.gz && \
+    tar -zxpf luarocks-${LUAROCKS_VER}.tar.gz && \
+    cd luarocks-${LUAROCKS_VER} && \
+    ./configure && make && make install && \
+    cd .. && \
+    rm luarocks-${LUAROCKS_VER}.tar.gz.asc && \
+    rm luarocks-${LUAROCKS_VER}.tar.gz && \
+    rm -rf luarocks-${LUAROCKS_VER}
+
+RUN luarocks install busted
+RUN luarocks install alt-getopt
+RUN luarocks install moonscript
+
+FROM ubuntu:24.04
+
+RUN apt-get update && \
+    apt-get install -y jq && \
+    rm -rf /var/lib/apt/lists/* && \
+    apt-get purge --auto-remove && \
+    apt-get clean
+
+COPY --from=builder /usr/local /usr/local
+
+COPY . /opt/test-runner
 WORKDIR /opt/test-runner
-COPY . .
-ENTRYPOINT ["/opt/test-runner/bin/run.sh"]
+ENTRYPOINT ["/opt/test-runner/bin/run.moon"]
diff --git a/README.md b/README.md
index 6636269..937970e 100644
--- a/README.md
+++ b/README.md
@@ -2,24 +2,14 @@
 
 The Docker image to automatically run tests on MoonScript solutions submitted to [Exercism].
 
-## Getting started
-
-Build the test runner, conforming to the [Test Runner interface specification](https://github.com/exercism/docs/blob/main/building/tooling/test-runners/interface.md).
-Update the files to match your track's needs. At the very least, you'll need to update `bin/run.sh`, `Dockerfile` and the test solutions in the `tests` directory
-
-- Tip: look for `TODO:` comments to point you towards code that need updating
-- Tip: look for `OPTIONAL:` comments to point you towards code that _could_ be useful
-- Tip: if it proves impossible for the Docker image to work on a read-only filesystem, remove the `--read-only` flag from the `bin/run-in-docker.sh` and `bin/run-tests-in-docker.sh` files.
-  We don't yet enforce a read-only file system in production, but we might in the future!
-
 ## Run the test runner
 
 To run the tests of a single solution, do the following:
 
 1. Open a terminal in the project's root
-2. Run `./bin/run.sh <exercise-slug> <solution-dir> <output-dir>`
+2. Run `bin/run.moon ${exercise_slug} ${solution_dir} ${output_dir}`
 
-Once the test runner has finished, its results will be written to `<output-dir>/results.json`.
+Once the test runner has finished, its results will be written to `${output_dir}/results.json`.
 
 ## Run the test runner on a solution using Docker
 
@@ -28,9 +18,9 @@ _This script is provided for testing purposes, as it mimics how test runners run
 To run the tests of a single solution using the Docker image, do the following:
 
 1. Open a terminal in the project's root
-2. Run `./bin/run-in-docker.sh <exercise-slug> <solution-dir> <output-dir>`
+2. Run `./bin/run-in-docker.sh ${exercise_slug} ${solution_dir} ${output_dir}`
 
-Once the test runner has finished, its results will be written to `<output-dir>/results.json`.
+Once the test runner has finished, its results will be written to `${output_dir}/results.json`.
 
 ## Run the tests
 
@@ -39,9 +29,9 @@ To run the tests to verify the behavior of the test runner, do the following:
 1. Open a terminal in the project's root
 2. Run `./bin/run-tests.sh`
 
-These are [golden tests][golden] that compare the `results.json` generated by running the current state of the code against the "known good" `tests/<test-name>/expected_results.json`. All files created during the test run itself are discarded.
+These are [golden tests][golden] that compare the `results.json` generated by running the current state of the code against the "known good" `tests/${test_name}/expected_results.json`. All files created during the test run itself are discarded.
 
-When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected `tests/<test-name>/expected_results.json` file(s).
+When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected `tests/${test_name}/expected_results.json` file(s).
 
 ## Run the tests using Docker
 
@@ -52,12 +42,14 @@ To run the tests to verify the behavior of the test runner using the Docker imag
 1. Open a terminal in the project's root
 2. Run `./bin/run-tests-in-docker.sh`
 
-These are [golden tests][golden] that compare the `results.json` generated by running the current state of the code against the "known good" `tests/<test-name>/expected_results.json`. All files created during the test run itself are discarded.
+These are [golden tests][golden] that compare the `results.json` generated by running the current state of the code against the "known good" `tests/${test_name}/expected_results.json`. All files created during the test run itself are discarded.
 
-When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected `tests/<test-name>/expected_results.json` file(s).
+When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected `tests/${test_name}/expected_results.json` file(s).
 
 ## Benchmarking
 
+**_NOTE: not implemented_**
+
 There are two scripts you can use to benchmark the test runner:
 
 1. `./bin/benchmark.sh`: benchmark the test runner code
diff --git a/bin/benchmark-in-docker.sh b/bin/benchmark-in-docker.sh
old mode 100755
new mode 100644
diff --git a/bin/benchmark.sh b/bin/benchmark.sh
old mode 100755
new mode 100644
diff --git a/bin/run-in-docker.sh b/bin/run-in-docker.sh
index 3672f7b..5c47d02 100755
--- a/bin/run-in-docker.sh
+++ b/bin/run-in-docker.sh
@@ -1,4 +1,5 @@
 #!/usr/bin/env sh
+set -e
 
 # Synopsis:
 # Run the test runner on a solution using the test runner Docker image.
@@ -6,22 +7,19 @@
 
 # Arguments:
 # $1: exercise slug
-# $2: path to solution folder
-# $3: path to output directory
+# $2: absolute path to solution folder
+# $3: absolute path to output directory
 
 # Output:
 # Writes the test results to a results.json file in the passed-in output directory.
 # The test results are formatted according to the specifications at https://github.com/exercism/docs/blob/main/building/tooling/test-runners/interface.md
 
 # Example:
-# ./bin/run-in-docker.sh two-fer path/to/solution/folder/ path/to/output/directory/
-
-# Stop executing when a command returns a non-zero return code
-set -e
+# ./bin/run-in-docker.sh two-fer /absolute/path/to/two-fer/solution/folder/ /absolute/path/to/output/directory/
 
 # If any required arguments is missing, print the usage and exit
 if [ -z "$1" ] || [ -z "$2" ] || [ -z "$3" ]; then
-    echo "usage: ./bin/run-in-docker.sh exercise-slug path/to/solution/folder/ path/to/output/directory/"
+    echo "usage: $0 exercise-slug /absolute/path/to/solution/folder/ /absolute/path/to/output/directory/"
     exit 1
 fi
 
@@ -43,4 +41,4 @@ docker run \
     --mount type=bind,src="${solution_dir}",dst=/solution \
     --mount type=bind,src="${output_dir}",dst=/output \
     --mount type=tmpfs,dst=/tmp \
-    exercism/moonscript-test-runner "${slug}" /solution /output 
+    exercism/moonscript-test-runner "${slug}" /solution /output
diff --git a/bin/run-tests-in-docker.sh b/bin/run-tests-in-docker.sh
index 1d4dcf9..1c38f02 100755
--- a/bin/run-tests-in-docker.sh
+++ b/bin/run-tests-in-docker.sh
@@ -1,4 +1,5 @@
 #!/usr/bin/env sh
+set -e
 
 # Synopsis:
 # Test the test runner Docker image by running it against a predefined set of 
@@ -12,9 +13,6 @@
 # Example:
 # ./bin/run-tests-in-docker.sh
 
-# Stop executing when a command returns a non-zero return code
-set -e
-
 # Build the Docker image
 docker build --rm -t exercism/moonscript-test-runner .
 
diff --git a/bin/run-tests.sh b/bin/run-tests.sh
index 2bb47b8..a4b7c07 100755
--- a/bin/run-tests.sh
+++ b/bin/run-tests.sh
@@ -1,4 +1,4 @@
-#!/usr/bin/env sh
+#!/usr/bin/env bash
 
 # Synopsis:
 # Test the test runner by running it against a predefined set of solutions 
@@ -8,30 +8,20 @@
 # Outputs the diff of the expected test results against the actual test results
 # generated by the test runner.
 
+# Exit status: the number of failed tests
+
 # Example:
 # ./bin/run-tests.sh
 
 exit_code=0
 
-# Iterate over all test directories
 for test_dir in tests/*; do
-    test_dir_name=$(basename "${test_dir}")
-    test_dir_path=$(realpath "${test_dir}")
-
-    bin/run.sh "${test_dir_name}" "${test_dir_path}" "${test_dir_path}"
-
-    # OPTIONAL: Normalize the results file
-    # If the results.json file contains information that changes between 
-    # different test runs (e.g. timing information or paths), you should normalize
-    # the results file to allow the diff comparison below to work as expected
-
-    file="results.json"
-    expected_file="expected_${file}"
-    echo "${test_dir_name}: comparing ${file} to ${expected_file}"
+    test_name=$(basename "${test_dir}")
+    test_path=$(realpath "${test_dir}")
 
-    if ! diff "${test_dir_path}/${file}" "${test_dir_path}/${expected_file}"; then
-        exit_code=1
-    fi
+    bin/run.moon "${test_name}" "${test_path}" "${test_path}" \
+    && bin/test-result-compare.lua "${test_dir}/results.json" "${test_dir}/expected_results.json" \
+    || (( ++exit_code ))
 done
 
 exit ${exit_code}
diff --git a/bin/run.moon b/bin/run.moon
new file mode 100755
index 0000000..8d78bb8
--- /dev/null
+++ b/bin/run.moon
@@ -0,0 +1,192 @@
+#! /usr/bin/env moon
+
+require 'moonscript'
+lfs = require 'lfs'
+json = (require 'dkjson').use_lpeg!
+getopt = require 'alt_getopt'
+local verbose
+
+import p from require 'moon'
+
+
+-- -----------------------------------------------------------
+show_help = (args) ->
+  print "Usage: #{args[0]} [-h] [-v]  slug  solution-dir  output-dir"
+  print "Where: -h   show this help"
+  print "       -v   verbose: show the output JSON"
+  os.exit!
+
+
+-- -----------------------------------------------------------
+file_exists = (path) ->
+  attrs = lfs.attributes path
+  not not attrs
+
+is_directory = (path) ->
+  attrs = lfs.attributes path
+  attrs and attrs.mode == 'directory'
+
+realpath = (path) ->
+  fh = io.popen "realpath #{path}"
+  dir = fh\read!
+  fh\close!
+  dir
+
+validate = (args) ->
+  show_help args unless #args == 3
+  {slug, src_dir, dest_dir} = args
+  assert slug != '', 'First arg, the slug, cannot be empty'
+  assert is_directory(src_dir), 'Second arg, the solution directory, must be a directory'
+  assert is_directory(dest_dir), 'Third arg, the output directory, must be a directory'
+
+  slug, realpath(src_dir), realpath(dest_dir)
+
+
+-- -----------------------------------------------------------
+run_tests = (slug, dir) ->
+  ok, err = lfs.chdir dir
+  assert ok, err
+
+  -- unskip tests
+  cmd = "perl -i.bak -pe 's{^\\s*\\Kpending\\b}{it}' *_spec.moon"
+  ok, result_type, status = os.execute cmd
+  assert ok
+
+  -- launch `busted`
+  fh = io.popen 'busted -o json', 'r'
+  json_output = fh\read 'a'
+  ok, exit_type, exit_status = fh\close!
+
+  if exit_type == 'signal'
+    return {
+      status: 'error',
+      message: json_output
+    }
+
+  data = json.decode json_output
+
+  if not data
+    return {
+      status: 'error',
+      message: json_output
+    }
+
+  if exit_status != 0 and #data.successes == 0 and #data.failures == 0 and #data.errors > 0
+    return {
+      status: 'error',
+      message: data.errors[1].message
+    }
+
+  results = {}
+
+  for test in *data.successes
+    results[test.element.name] = {
+      status: 'pass',
+      name: test.element.name,
+    }
+
+  for test in *data.failures
+    results[test.element.name] = {
+      status: 'fail',
+      name: test.element.name,
+      message: test.trace.message,
+    }
+
+  results
+
+
+-- -----------------------------------------------------------
+get_test_bodies = (slug, dir) ->
+  ok, err = lfs.chdir dir
+  assert ok, err
+
+  order = {}
+  bodies = {}
+
+  test_file = "#{slug\gsub('-', '_')}_spec.moon"
+  return unless file_exists test_file -- let `busted` handle the error messaging
+
+  fh = io.open test_file, 'r'
+
+  pattern = (word) -> '^%s+' .. word .. '%s+[\'"](.+)[\'"],%s+->'
+  patterns = it: pattern('it'), pending: pattern('pending')
+
+  local test_name
+  test_body = {}
+  in_test = false
+
+  for line in fh\lines!
+    if line\match '^%s+describe '
+      if test_name
+        bodies[test_name] = table.concat test_body, '\n'
+        test_body = {}
+        test_name = nil
+      in_test = false
+
+    m = line\match(patterns.it) or line\match(patterns.pending)
+    if not m
+      if in_test
+        table.insert test_body, line
+    else
+      table.insert order, m
+      if in_test
+        bodies[test_name] = table.concat test_body, '\n'
+        test_body = {}
+      test_name = m
+      in_test = true
+
+  fh\close!
+  bodies[test_name] = table.concat test_body, '\n'
+  order, bodies
+
+
+-- -----------------------------------------------------------
+write_results = (slug, test_results, names, bodies, dir) ->
+  ok, err = lfs.chdir dir
+  assert ok, "#{err}: #{dir}"
+  
+  results = version: 2, status: nil, tests: {}
+
+  if test_results.status
+    -- this was an error result
+    results.status = test_results.status
+    results.message = test_results.message
+
+  else
+    status = 'pass'
+    for name in *names
+      test = test_results[name]
+      assert test, "no test result for #{name}"
+      status = 'fail' if test.status == 'fail'
+      test.test_code = bodies[name]
+      table.insert results.tests, test
+    results.status = status
+
+  fh = io.open 'results.json', 'w'
+  fh\write (json.encode results) .. '\n'
+  fh\close!
+
+  os.execute "jq . results.json" if verbose
+
+
+-- -----------------------------------------------------------
+main = (args) ->
+  opts, optind = getopt.get_opts args, 'hv', {}
+
+  show_help args if opts.h
+  verbose = not not opts.v
+  table.remove args, 1 for _ = 1, optind - 1
+
+  slug, src_dir, dest_dir = validate args
+
+  print "#{slug}: testing ..."
+
+  test_names_ordered, test_code = get_test_bodies slug, src_dir
+
+  test_results = run_tests slug, src_dir
+
+  write_results slug, test_results, test_names_ordered, test_code, dest_dir
+
+  print "#{slug}: ... done"
+
+main arg
diff --git a/bin/run.sh b/bin/run.sh
deleted file mode 100755
index f1e8b78..0000000
--- a/bin/run.sh
+++ /dev/null
@@ -1,60 +0,0 @@
-#!/usr/bin/env sh
-
-# Synopsis:
-# Run the test runner on a solution.
-
-# Arguments:
-# $1: exercise slug
-# $2: path to solution folder
-# $3: path to output directory
-
-# Output:
-# Writes the test results to a results.json file in the passed-in output directory.
-# The test results are formatted according to the specifications at https://github.com/exercism/docs/blob/main/building/tooling/test-runners/interface.md
-
-# Example:
-# ./bin/run.sh two-fer path/to/solution/folder/ path/to/output/directory/
-
-# If any required arguments is missing, print the usage and exit
-if [ -z "$1" ] || [ -z "$2" ] || [ -z "$3" ]; then
-    echo "usage: ./bin/run.sh exercise-slug path/to/solution/folder/ path/to/output/directory/"
-    exit 1
-fi
-
-slug="$1"
-solution_dir=$(realpath "${2%/}")
-output_dir=$(realpath "${3%/}")
-results_file="${output_dir}/results.json"
-
-# Create the output directory if it doesn't exist
-mkdir -p "${output_dir}"
-
-echo "${slug}: testing..."
-
-# Run the tests for the provided implementation file and redirect stdout and
-# stderr to capture it
-test_output=$(false)
-# TODO: substitute "false" with the actual command to run the test:
-# test_output=$(command_to_run_tests 2>&1)
-
-# Write the results.json file based on the exit code of the command that was 
-# just executed that tested the implementation file
-if [ $? -eq 0 ]; then
-    jq -n '{version: 1, status: "pass"}' > ${results_file}
-else
-    # OPTIONAL: Sanitize the output
-    # In some cases, the test output might be overly verbose, in which case stripping
-    # the unneeded information can be very helpful to the student
-    # sanitized_test_output=$(printf "${test_output}" | sed -n '/Test results:/,$p')
-
-    # OPTIONAL: Manually add colors to the output to help scanning the output for errors
-    # If the test output does not contain colors to help identify failing (or passing)
-    # tests, it can be helpful to manually add colors to the output
-    # colorized_test_output=$(echo "${test_output}" \
-    #      | GREP_COLOR='01;31' grep --color=always -E -e '^(ERROR:.*|.*failed)$|$' \
-    #      | GREP_COLOR='01;32' grep --color=always -E -e '^.*passed$|$')
-
-    jq -n --arg output "${test_output}" '{version: 1, status: "fail", message: $output}' > ${results_file}
-fi
-
-echo "${slug}: done"
diff --git a/bin/test-result-compare.lua b/bin/test-result-compare.lua
new file mode 100755
index 0000000..2dd8c60
--- /dev/null
+++ b/bin/test-result-compare.lua
@@ -0,0 +1,25 @@
+#!/usr/bin/env lua
+
+local json = require('dkjson')
+local tablex = require('pl.tablex')
+local pretty = require('pl.pretty')
+
+local function load_json(file)
+    local fd <close> = io.open(file)
+    return json.decode(fd:read('a'))
+end
+
+local result = load_json(arg[1])
+local expected = load_json(arg[2])
+
+if not tablex.deepcompare(result, expected) then
+    print('\n========[  RESULT  ]========\n')
+    print(pretty.write(result))
+
+    print('\n========[ EXPECTED ]========\n')
+    print(pretty.write(expected))
+
+    os.exit(1)
+end
+
+os.exit(0)
diff --git a/bin/validate-track-in-docker b/bin/validate-track-in-docker
new file mode 100755
index 0000000..f2c4b7b
--- /dev/null
+++ b/bin/validate-track-in-docker
@@ -0,0 +1,33 @@
+#!/usr/bin/env sh
+
+# Synopsis:
+# Test the test runner Docker image by running it against the example solutions
+# from the language track.  The test runner Docker image is built
+# automatically.  The track is cloned from git into a subdirectory.
+
+# Output:
+# The actual test results generated by the test runner Docker image.
+
+# Example:
+# ./bin/validate-track-in-docker.sh
+
+# Stop executing when a command returns a non-zero return code
+set -e
+
+# Fetch the moonscript track
+rm -rf track/
+git clone --depth 1 https://github.com/exercism/moonscript.git track
+
+# Build the Docker image
+docker build --rm -t exercism/moonscript-test-runner .
+
+# Run the Docker image using the settings mimicking the production environment
+docker run \
+    --rm \
+    --network none \
+    --read-only \
+    --mount type=bind,src="${PWD}/track",dst=/opt/test-runner/track \
+    --mount type=tmpfs,dst=/tmp \
+    --workdir /opt/test-runner \
+    --entrypoint /opt/test-runner/track/bin/verify-exercises \
+    exercism/moonscript-test-runner
diff --git a/tests/.gitignore b/tests/.gitignore
new file mode 100644
index 0000000..e345bae
--- /dev/null
+++ b/tests/.gitignore
@@ -0,0 +1,2 @@
+*.bak
+results.json
diff --git a/tests/all-fail/expected_results.json b/tests/all-fail/expected_results.json
deleted file mode 100644
index 9ef8b6f..0000000
--- a/tests/all-fail/expected_results.json
+++ /dev/null
@@ -1,5 +0,0 @@
-{
-  "version": 1,
-  "status": "fail",
-  "message": "TODO: replace with correct output"
-}
diff --git a/tests/empty-file/expected_results.json b/tests/empty-file/expected_results.json
deleted file mode 100644
index 9ef8b6f..0000000
--- a/tests/empty-file/expected_results.json
+++ /dev/null
@@ -1,5 +0,0 @@
-{
-  "version": 1,
-  "status": "fail",
-  "message": "TODO: replace with correct output"
-}
diff --git a/tests/example-all-fail/.busted b/tests/example-all-fail/.busted
new file mode 100644
index 0000000..86b84e7
--- /dev/null
+++ b/tests/example-all-fail/.busted
@@ -0,0 +1,5 @@
+return {
+  default = {
+    ROOT = { '.' }
+  }
+}
diff --git a/tests/example-all-fail/example_all_fail.moon b/tests/example-all-fail/example_all_fail.moon
new file mode 100644
index 0000000..a8420de
--- /dev/null
+++ b/tests/example-all-fail/example_all_fail.moon
@@ -0,0 +1,4 @@
+leap_year = (number) ->
+    number % 2 == 1
+
+leap_year
diff --git a/tests/example-all-fail/example_all_fail_spec.moon b/tests/example-all-fail/example_all_fail_spec.moon
new file mode 100644
index 0000000..98e9954
--- /dev/null
+++ b/tests/example-all-fail/example_all_fail_spec.moon
@@ -0,0 +1,8 @@
+is_leap_year = require 'example_all_fail'
+
+describe 'leap', ->
+  it 'a known leap year', ->
+    assert.is_true is_leap_year 1996
+
+  it 'any old year', ->
+    assert.is_false is_leap_year 1997
diff --git a/tests/example-all-fail/expected_results.json b/tests/example-all-fail/expected_results.json
new file mode 100644
index 0000000..b9866f8
--- /dev/null
+++ b/tests/example-all-fail/expected_results.json
@@ -0,0 +1 @@
+{"tests":[{"message":"Expected objects to be the same.\nPassed in:\n(boolean) false\nExpected:\n(boolean) true","test_code":"    assert.is_true is_leap_year 1996\n","name":"a known leap year","status":"fail"},{"message":"Expected objects to be the same.\nPassed in:\n(boolean) true\nExpected:\n(boolean) false","test_code":"    assert.is_false is_leap_year 1997","name":"any old year","status":"fail"}],"version":2,"status":"fail"}
diff --git a/tests/example-empty-file/.busted b/tests/example-empty-file/.busted
new file mode 100644
index 0000000..86b84e7
--- /dev/null
+++ b/tests/example-empty-file/.busted
@@ -0,0 +1,5 @@
+return {
+  default = {
+    ROOT = { '.' }
+  }
+}
diff --git a/tests/example-empty-file/example_empty_file.moon b/tests/example-empty-file/example_empty_file.moon
new file mode 100644
index 0000000..e69de29
diff --git a/tests/example-empty-file/example_empty_file_spec.moon b/tests/example-empty-file/example_empty_file_spec.moon
new file mode 100644
index 0000000..aa8e1f9
--- /dev/null
+++ b/tests/example-empty-file/example_empty_file_spec.moon
@@ -0,0 +1,14 @@
+is_leap_year = require 'example_empty_file'
+
+describe 'leap', ->
+  it 'a known leap year', ->
+    assert.is_true is_leap_year 1996
+
+  it 'any old year', ->
+    assert.is_false is_leap_year 1997
+
+  it 'turn of the 20th century', ->
+    assert.is_false is_leap_year 1900
+
+  it 'turn of the 21st century', ->
+    assert.is_true is_leap_year 2400
diff --git a/tests/example-empty-file/expected_results.json b/tests/example-empty-file/expected_results.json
new file mode 100644
index 0000000..a15497a
--- /dev/null
+++ b/tests/example-empty-file/expected_results.json
@@ -0,0 +1 @@
+{"status":"error","tests":[],"message":"./example_empty_file_spec.moon:5: attempt to call a boolean value (upvalue 'is_leap_year')","version":2}
diff --git a/tests/example-partial-fail/.busted b/tests/example-partial-fail/.busted
new file mode 100644
index 0000000..86b84e7
--- /dev/null
+++ b/tests/example-partial-fail/.busted
@@ -0,0 +1,5 @@
+return {
+  default = {
+    ROOT = { '.' }
+  }
+}
diff --git a/tests/example-partial-fail/example_partial_fail.moon b/tests/example-partial-fail/example_partial_fail.moon
new file mode 100644
index 0000000..c338f3e
--- /dev/null
+++ b/tests/example-partial-fail/example_partial_fail.moon
@@ -0,0 +1,7 @@
+leap_year = (number) ->
+  is_divisible_by = (a_number) ->
+    number % a_number == 0
+
+  is_divisible_by(401) or (is_divisible_by(4) and (not is_divisible_by(100)))
+
+leap_year
diff --git a/tests/example-partial-fail/example_partial_fail_spec.moon b/tests/example-partial-fail/example_partial_fail_spec.moon
new file mode 100644
index 0000000..02de8d1
--- /dev/null
+++ b/tests/example-partial-fail/example_partial_fail_spec.moon
@@ -0,0 +1,14 @@
+is_leap_year = require 'example_partial_fail'
+
+describe 'leap', ->
+  it 'a known leap year', ->
+    assert.is_true is_leap_year 1996
+
+  it 'any old year', ->
+    assert.is_false is_leap_year 1997
+
+  it 'turn of the 20th century', ->
+    assert.is_false is_leap_year 1900
+
+  it 'turn of the 21st century', ->
+    assert.is_true is_leap_year 2400
diff --git a/tests/example-partial-fail/expected_results.json b/tests/example-partial-fail/expected_results.json
new file mode 100644
index 0000000..09dee9e
--- /dev/null
+++ b/tests/example-partial-fail/expected_results.json
@@ -0,0 +1 @@
+{"status":"fail","tests":[{"test_code":"    assert.is_true is_leap_year 1996\n","status":"pass","name":"a known leap year"},{"test_code":"    assert.is_false is_leap_year 1997\n","status":"pass","name":"any old year"},{"test_code":"    assert.is_false is_leap_year 1900\n","status":"pass","name":"turn of the 20th century"},{"test_code":"    assert.is_true is_leap_year 2400","message":"Expected objects to be the same.\nPassed in:\n(boolean) false\nExpected:\n(boolean) true","name":"turn of the 21st century","status":"fail"}],"version":2}
diff --git a/tests/example-success/.busted b/tests/example-success/.busted
new file mode 100644
index 0000000..86b84e7
--- /dev/null
+++ b/tests/example-success/.busted
@@ -0,0 +1,5 @@
+return {
+  default = {
+    ROOT = { '.' }
+  }
+}
diff --git a/tests/example-success/example_success.moon b/tests/example-success/example_success.moon
new file mode 100644
index 0000000..47d7028
--- /dev/null
+++ b/tests/example-success/example_success.moon
@@ -0,0 +1,7 @@
+leap_year = (number) ->
+  is_divisible_by = (a_number) ->
+    number % a_number == 0
+
+  is_divisible_by(400) or (is_divisible_by(4) and (not is_divisible_by(100)))
+
+leap_year
diff --git a/tests/example-success/example_success_spec.moon b/tests/example-success/example_success_spec.moon
new file mode 100644
index 0000000..8127303
--- /dev/null
+++ b/tests/example-success/example_success_spec.moon
@@ -0,0 +1,26 @@
+is_leap_year = require 'example_success'
+
+describe 'leap', ->
+  it 'a known leap year', ->
+    assert.is_true is_leap_year 1996
+
+  it 'any old year', ->
+    assert.is_false is_leap_year 1997
+
+  it 'turn of the 20th century', ->
+    assert.is_false is_leap_year 1900
+
+  it 'turn of the 21st century', ->
+    assert.is_true is_leap_year 2400
+
+  it 'handles tests with parens in strings', ->
+    s = ')'
+    assert.is_true true
+
+  it "handles test names with 'apostrophes'", ->
+    assert.is_true true
+
+  it 'handles tests with multiple lines in the body', ->
+    assert.is_false false
+
+    assert.is_true true
diff --git a/tests/example-success/expected_results.json b/tests/example-success/expected_results.json
new file mode 100644
index 0000000..be1a520
--- /dev/null
+++ b/tests/example-success/expected_results.json
@@ -0,0 +1 @@
+{"tests":[{"name":"a known leap year","status":"pass","test_code":"    assert.is_true is_leap_year 1996\n"},{"name":"any old year","status":"pass","test_code":"    assert.is_false is_leap_year 1997\n"},{"name":"turn of the 20th century","status":"pass","test_code":"    assert.is_false is_leap_year 1900\n"},{"name":"turn of the 21st century","status":"pass","test_code":"    assert.is_true is_leap_year 2400\n"},{"name":"handles tests with parens in strings","status":"pass","test_code":"    s = ')'\n    assert.is_true true\n"},{"name":"handles test names with 'apostrophes'","status":"pass","test_code":"    assert.is_true true\n"},{"name":"handles tests with multiple lines in the body","status":"pass","test_code":"    assert.is_false false\n\n    assert.is_true true"}],"status":"pass","version":2}
diff --git a/tests/example-syntax-error/.busted b/tests/example-syntax-error/.busted
new file mode 100644
index 0000000..86b84e7
--- /dev/null
+++ b/tests/example-syntax-error/.busted
@@ -0,0 +1,5 @@
+return {
+  default = {
+    ROOT = { '.' }
+  }
+}
diff --git a/tests/example-syntax-error/example_syntax_error.moon b/tests/example-syntax-error/example_syntax_error.moon
new file mode 100644
index 0000000..9f83213
--- /dev/null
+++ b/tests/example-syntax-error/example_syntax_error.moon
@@ -0,0 +1 @@
+leap_year =
diff --git a/tests/example-syntax-error/example_syntax_error_spec.moon b/tests/example-syntax-error/example_syntax_error_spec.moon
new file mode 100644
index 0000000..cc23079
--- /dev/null
+++ b/tests/example-syntax-error/example_syntax_error_spec.moon
@@ -0,0 +1,8 @@
+is_leap_year = require 'example_syntax_error'
+
+describe 'leap', ->
+  it 'a known leap year', ->
+    assert.is_true is_leap_year 1996
+
+  it 'any old year', ->
+    assert.is_false is_leap_year 1997
diff --git a/tests/example-syntax-error/expected_results.json b/tests/example-syntax-error/expected_results.json
new file mode 100644
index 0000000..be52766
--- /dev/null
+++ b/tests/example-syntax-error/expected_results.json
@@ -0,0 +1 @@
+{"tests":[],"version":2,"message":"Failed to encode test results to json: type 'function' is not supported by JSON.\n","status":"error"}
diff --git a/tests/partial-fail/expected_results.json b/tests/partial-fail/expected_results.json
deleted file mode 100644
index 9ef8b6f..0000000
--- a/tests/partial-fail/expected_results.json
+++ /dev/null
@@ -1,5 +0,0 @@
-{
-  "version": 1,
-  "status": "fail",
-  "message": "TODO: replace with correct output"
-}
diff --git a/tests/success/expected_results.json b/tests/success/expected_results.json
deleted file mode 100644
index 6c2223e..0000000
--- a/tests/success/expected_results.json
+++ /dev/null
@@ -1,4 +0,0 @@
-{
-  "version": 1,
-  "status": "pass"
-}
diff --git a/tests/syntax-error/expected_results.json b/tests/syntax-error/expected_results.json
deleted file mode 100644
index afc4d4e..0000000
--- a/tests/syntax-error/expected_results.json
+++ /dev/null
@@ -1,5 +0,0 @@
-{
-  "version": 1,
-  "status": "error",
-  "message": "TODO: replace with correct output"
-}