diff --git a/.github/workflows/hil.yml b/.github/workflows/hil.yml
index 79bbe6f44..c4665f313 100644
--- a/.github/workflows/hil.yml
+++ b/.github/workflows/hil.yml
@@ -52,8 +52,7 @@ jobs:
           components: rust-src
 
       - name: Run tests
-        working-directory: hil-test
-        run: cargo ${{ matrix.target.soc }}
+        run: cargo xtask run-tests ${{ matrix.target.soc }}
 
   # Test Xtensa targets:
   # TODO: Add jobs for Xtensa once supported by `probe-rs`
diff --git a/hil-test/tests/aes.rs b/hil-test/tests/aes.rs
index 8cb6735a6..139772dbb 100644
--- a/hil-test/tests/aes.rs
+++ b/hil-test/tests/aes.rs
@@ -1,5 +1,7 @@
 //! AES Test
 
+//% CHIPS: esp32 esp32c3 esp32c6 esp32h2 esp32s2 esp32s3
+
 #![no_std]
 #![no_main]
 
@@ -23,30 +25,6 @@ impl Context<'_> {
     }
 }
 
-#[cfg(not(any(
-    feature = "esp32c3",
-    feature = "esp32c6",
-    feature = "esp32h2",
-    feature = "esp32s3"
-)))]
-mod not_test {
-    #[esp_hal::entry]
-    fn main() -> ! {
-        semihosting::process::exit(0)
-    }
-    #[panic_handler]
-    fn panic(_info: &core::panic::PanicInfo) -> ! {
-        loop {}
-    }
-}
-
-#[cfg(test)]
-#[cfg(any(
-    feature = "esp32c3",
-    feature = "esp32c6",
-    feature = "esp32h2",
-    feature = "esp32s3"
-))]
 #[embedded_test::tests]
 mod tests {
     use defmt::assert_eq;
diff --git a/xtask/src/lib.rs b/xtask/src/lib.rs
index 3e3448ff9..58bb02374 100644
--- a/xtask/src/lib.rs
+++ b/xtask/src/lib.rs
@@ -296,27 +296,33 @@ pub fn run_example(
         log::info!("  Features: {}", example.features().join(","));
     }
 
-    let bin = if example
-        .example_path()
-        .strip_prefix(package_path)?
-        .starts_with("src/bin")
-    {
-        format!("--bin={}", example.name())
+    let package = example.example_path().strip_prefix(package_path)?;
+    log::info!("Package: {:?}", package);
+    let (bin, subcommand) = if package.starts_with("src/bin") {
+        (format!("--bin={}", example.name()), "run")
+    } else if package.starts_with("tests") {
+        (format!("--test={}", example.name()), "test")
     } else {
-        format!("--example={}", example.name())
+        (format!("--example={}", example.name()), "run")
     };
 
     let mut features = example.features().to_vec();
     features.push(chip.to_string());
 
     let mut builder = CargoArgsBuilder::default()
-        .subcommand("run")
+        .subcommand(subcommand)
         .arg("-Zbuild-std=alloc,core")
         .arg("--release")
         .target(target)
         .features(&features)
         .arg(bin);
 
+    // probe-rs cannot currently do auto detection, so we need to tell probe-rs run
+    // which chip we are testing
+    if subcommand == "test" {
+        builder = builder.arg("--").arg("--chip").arg(format!("{}", chip));
+    }
+
     // If targeting an Xtensa device, we must use the '+esp' toolchain modifier:
     if target.starts_with("xtensa") {
         builder = builder.toolchain("esp");
diff --git a/xtask/src/main.rs b/xtask/src/main.rs
index 7007eb9db..5d5d864a9 100644
--- a/xtask/src/main.rs
+++ b/xtask/src/main.rs
@@ -25,6 +25,8 @@ enum Cli {
     GenerateEfuseFields(GenerateEfuseFieldsArgs),
     /// Run the given example for the specified chip.
     RunExample(RunExampleArgs),
+    /// Run all applicable tests for a specified chip.
+    RunTests(RunTestsArgs),
 }
 
 #[derive(Debug, Args)]
@@ -100,6 +102,13 @@ struct RunExampleArgs {
     example: String,
 }
 
+#[derive(Debug, Args)]
+struct RunTestsArgs {
+    /// Which chip to run the examples for.
+    #[arg(value_enum)]
+    chip: Chip,
+}
+
 // ----------------------------------------------------------------------------
 // Application
 
@@ -118,6 +127,7 @@ fn main() -> Result<()> {
         Cli::BumpVersion(args) => bump_version(&workspace, args),
         Cli::GenerateEfuseFields(args) => generate_efuse_src(&workspace, args),
         Cli::RunExample(args) => run_example(&workspace, args),
+        Cli::RunTests(args) => run_tests(&workspace, args),
     }
 }
 
@@ -293,11 +303,10 @@ fn run_example(workspace: &Path, mut args: RunExampleArgs) -> Result<()> {
     // Absolute path of the package's root:
     let package_path = xtask::windows_safe_path(&workspace.join(args.package.to_string()));
 
-    // Absolute path to the directory containing the examples:
-    let example_path = if args.package == Package::Examples {
-        package_path.join("src").join("bin")
-    } else {
-        package_path.join("examples")
+    let example_path = match args.package {
+        Package::Examples => package_path.join("src").join("bin"),
+        Package::HilTest => package_path.join("tests"),
+        _ => package_path.join("examples"),
     };
 
     // Determine the appropriate build target for the given package and chip:
@@ -325,6 +334,26 @@ fn run_example(workspace: &Path, mut args: RunExampleArgs) -> Result<()> {
     Ok(())
 }
 
+fn run_tests(workspace: &PathBuf, args: RunTestsArgs) -> Result<(), anyhow::Error> {
+    // Absolute path of the package's root:
+    let package_path = xtask::windows_safe_path(&workspace.join("hil-test"));
+
+    // Determine the appropriate build target for the given package and chip:
+    let target = target_triple(&Package::HilTest, &args.chip)?;
+
+    // Load all examples and parse their metadata:
+    let tests = xtask::load_examples(&package_path.join("tests"))?;
+    let tests = tests.iter()
+        // Filter down the examples to only those for which the specified chip is supported:
+        .filter(|example| example.supports_chip(args.chip));
+
+    for test in tests {
+        xtask::run_example(&package_path, args.chip, target, &test)?;
+    }
+
+    Ok(())
+}
+
 // ----------------------------------------------------------------------------
 // Helper Functions