diff --git a/Dockerfile b/Dockerfile
index ec8c704..6cfc174 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -2,29 +2,35 @@ ARG REPO=alpine
 ARG IMAGE=3.18
 FROM ${REPO}:${IMAGE} AS builder
 
+# set up Scarb
 ARG VERSION=v2.6.5
 ARG RELEASE=scarb-${VERSION}-x86_64-unknown-linux-musl
 
-RUN apk add --no-cache curl
-
-RUN mkdir opt/test-runner
-RUN mkdir opt/test-runner/bin
-WORKDIR /tmp
+WORKDIR /opt/test-runner/bin/scarb
 ADD https://github.com/software-mansion/scarb/releases/download/${VERSION}/${RELEASE}.tar.gz .
-RUN tar -xf ${RELEASE}.tar.gz \
-    && rm -rf /tmp/${RELEASE}/doc \
-    && mv /tmp/${RELEASE} /opt/test-runner/bin/scarb
+RUN tar -xf ${RELEASE}.tar.gz --strip-components=1 \
+    && rm -rf ./doc \
+    ./bin/scarb-cairo-language-server \
+    ./bin/scarb-cairo-run \
+    ./bin/scarb-snforge-test-collector
 
 FROM ${REPO}:${IMAGE} AS runner
 
-# install packages required to run the tests
-# hadolint ignore=DL3018
+# install jq package to format test results
 RUN apk add --no-cache jq
 
 COPY --from=builder /opt/test-runner/bin/scarb /opt/test-runner/bin/scarb
 ENV PATH=$PATH:/opt/test-runner/bin/scarb/bin
 
 WORKDIR /opt/test-runner
-COPY . .
-# Initialize a scarb cache
+
+COPY bin/run.sh bin/run.sh
+
+# Initialize a Cairo cache
+RUN mkdir -p init-cairo-cache/src
+COPY Scarb.toml init-cairo-cache/
+RUN echo '// dummy file' > init-cairo-cache/src/lib.cairo \
+    && scarb --manifest-path init-cairo-cache/Scarb.toml --release build \
+    && rm -rf init-cairo-cache/
+
 ENTRYPOINT ["/opt/test-runner/bin/run.sh"]
diff --git a/Scarb.toml b/Scarb.toml
new file mode 100644
index 0000000..0f2f16b
--- /dev/null
+++ b/Scarb.toml
@@ -0,0 +1,4 @@
+[package]
+name = "leap"
+version = "0.1.0"
+edition = "2023_11"
diff --git a/bin/run.sh b/bin/run.sh
index 7061ac4..323a07b 100755
--- a/bin/run.sh
+++ b/bin/run.sh
@@ -46,16 +46,13 @@ cd "${start_dir}" || exit 1
 if [ ${exit_code} -eq 0 ]; then
     jq -n '{version: 1, status: "pass"}' >"${results_file}"
 else
-    # Sanitize the output
-    test_output_inline=$(printf '%s' "$test_output")
-
     # Try to distinguish between failing tests and errors
-    if echo "$test_output_inline" | grep -q "error:"; then
+    if echo "$test_output" | grep -q "error:"; then
         status="error"
-        sanitized_test_output=$(echo "$test_output_inline" | sed '/Compiling.*$/d' | sed -n -e '/error: could not compile/q;p' | sed "s@$solution_dir@@g")
+        sanitized_test_output=$(echo "$test_output" | sed -n "/Compiling.*$/d ; s@$solution_dir@@g ; /error: could not compile/q;p")
     else
         status="fail"
-        sanitized_test_output=$(echo "$test_output_inline" | awk '/failures:/{y=1;next}y' | sed -n -e '/Error: test result/q;p' | sed -r 's/   //g')
+        sanitized_test_output=$(echo "$test_output" | sed -n '1,/failures:/d ; /Error: test result/q;p' | sed 's/[[:space:]]\{3\}.\+:://g ; /./G')
     fi
 
     jq -n --arg output "${sanitized_test_output}" --arg status "${status}" '{version: 1, status: $status, message: $output}' >"${results_file}"
diff --git a/tests/all-fail/expected_results.json b/tests/all-fail/expected_results.json
index c74fcd5..776cd16 100644
--- a/tests/all-fail/expected_results.json
+++ b/tests/all-fail/expected_results.json
@@ -1,5 +1,5 @@
 {
   "version": 1,
   "status": "fail",
-  "message": "leap_leap::leap::year_divisible_by_400_is_leap_year - Panicked with \"assertion failed: `is_leap_year(2000)`.\".\nleap_leap::leap::year_divisible_by_4_and_5_is_still_a_leap_year - Panicked with \"assertion failed: `is_leap_year(1960)`.\".\nleap_leap::leap::year_divisible_by_4_not_divisible_by_100_in_leap_year - Panicked with \"assertion failed: `is_leap_year(1996)`.\".\nleap_leap::leap::year_divisible_by_2_not_divisible_by_4_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1970)`.\".\nleap_leap::leap::year_divisible_by_400_but_not_by_125_is_still_a_leap_year - Panicked with \"assertion failed: `is_leap_year(2400)`.\".\nleap_leap::leap::year_not_divisible_by_4_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2015)`.\".\nleap_leap::leap::year_divisible_by_200_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1800)`.\".\nleap_leap::leap::year_divisible_by_100_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2100)`.\".\nleap_leap::leap::year_divisible_by_100_but_not_by_3_is_still_not_a_leap_year - Panicked with \"assertion failed: `!is_leap_year(1900)`.\"."
+  "message": "year_divisible_by_400_is_leap_year - Panicked with \"assertion failed: `is_leap_year(2000)`.\".\n\nyear_divisible_by_4_and_5_is_still_a_leap_year - Panicked with \"assertion failed: `is_leap_year(1960)`.\".\n\nyear_divisible_by_4_not_divisible_by_100_in_leap_year - Panicked with \"assertion failed: `is_leap_year(1996)`.\".\n\nyear_divisible_by_2_not_divisible_by_4_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1970)`.\".\n\nyear_divisible_by_400_but_not_by_125_is_still_a_leap_year - Panicked with \"assertion failed: `is_leap_year(2400)`.\".\n\nyear_not_divisible_by_4_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2015)`.\".\n\nyear_divisible_by_200_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1800)`.\".\n\nyear_divisible_by_100_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2100)`.\".\n\nyear_divisible_by_100_but_not_by_3_is_still_not_a_leap_year - Panicked with \"assertion failed: `!is_leap_year(1900)`.\"."
 }
diff --git a/tests/partial-fail/expected_results.json b/tests/partial-fail/expected_results.json
index 2d41d68..fde3de2 100644
--- a/tests/partial-fail/expected_results.json
+++ b/tests/partial-fail/expected_results.json
@@ -1,5 +1,5 @@
 {
   "version": 1,
   "status": "fail",
-  "message": "leap_leap::leap::year_divisible_by_100_but_not_by_3_is_still_not_a_leap_year - Panicked with \"assertion failed: `!is_leap_year(1900)`.\".\nleap_leap::leap::year_divisible_by_100_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2100)`.\".\nleap_leap::leap::year_divisible_by_200_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1800)`.\"."
+  "message": "year_divisible_by_100_but_not_by_3_is_still_not_a_leap_year - Panicked with \"assertion failed: `!is_leap_year(1900)`.\".\n\nyear_divisible_by_100_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(2100)`.\".\n\nyear_divisible_by_200_not_divisible_by_400_in_common_year - Panicked with \"assertion failed: `!is_leap_year(1800)`.\"."
 }