diff --git a/tests/integration/coverage_tests.rs b/tests/integration/coverage_tests.rs
index 5353996a0d..1bc3e887a8 100644
--- a/tests/integration/coverage_tests.rs
+++ b/tests/integration/coverage_tests.rs
@@ -3,7 +3,6 @@
 use deno_core::serde_json;
 use std::fs;
 use test_util as util;
-use test_util::itest;
 use test_util::TempDir;
 use util::assert_starts_with;
 use util::env_vars_for_npm_tests;
@@ -631,9 +630,3 @@ File         | Branch % | Line % |
 ",
   );
 }
-
-itest!(no_files_found {
-  args: "coverage doesnt_exist.js",
-  exit_code: 1,
-  output: "coverage/doesnt_exist.out",
-});
diff --git a/tests/integration/flags_tests.rs b/tests/integration/flags_tests.rs
index c898c199cc..751f4aafe4 100644
--- a/tests/integration/flags_tests.rs
+++ b/tests/integration/flags_tests.rs
@@ -1,21 +1,8 @@
 // Copyright 2018-2024 the Deno authors. All rights reserved. MIT license.
 
 use test_util as util;
-use test_util::itest;
 use util::assert_contains;
 
-#[test]
-fn help_flag() {
-  let status = util::deno_cmd()
-    .current_dir(util::testdata_path())
-    .arg("--help")
-    .spawn()
-    .unwrap()
-    .wait()
-    .unwrap();
-  assert!(status.success());
-}
-
 #[test]
 fn help_output() {
   let output = util::deno_cmd()
@@ -56,32 +43,3 @@ fn help_output() {
     assert_contains!(stdout, description);
   }
 }
-
-#[test]
-fn version_short_flag() {
-  let status = util::deno_cmd()
-    .current_dir(util::testdata_path())
-    .arg("-V")
-    .spawn()
-    .unwrap()
-    .wait()
-    .unwrap();
-  assert!(status.success());
-}
-
-#[test]
-fn version_long_flag() {
-  let status = util::deno_cmd()
-    .current_dir(util::testdata_path())
-    .arg("--version")
-    .spawn()
-    .unwrap()
-    .wait()
-    .unwrap();
-  assert!(status.success());
-}
-
-itest!(types {
-  args: "types",
-  output: "types/types.out",
-});
diff --git a/tests/integration/jsr_tests.rs b/tests/integration/jsr_tests.rs
index f78bea92d9..307398fc54 100644
--- a/tests/integration/jsr_tests.rs
+++ b/tests/integration/jsr_tests.rs
@@ -4,87 +4,11 @@ use deno_core::serde_json::json;
 use deno_core::serde_json::Value;
 use deno_lockfile::Lockfile;
 use test_util as util;
-use test_util::itest;
 use url::Url;
 use util::assert_contains;
 use util::assert_not_contains;
-use util::env_vars_for_jsr_npm_tests;
-use util::env_vars_for_jsr_tests;
 use util::TestContextBuilder;
 
-itest!(no_module_graph_run {
-  args: "run jsr/no_module_graph/main.ts",
-  output: "jsr/no_module_graph/main.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(no_module_graph_info {
-  args: "info jsr/no_module_graph/main.ts",
-  output: "jsr/no_module_graph/main_info.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(same_package_multiple_versions {
-  args: "run --quiet jsr/no_module_graph/multiple.ts",
-  output: "jsr/no_module_graph/multiple.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(module_graph_run {
-  args: "run jsr/module_graph/main.ts",
-  output: "jsr/module_graph/main.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(module_graph_info {
-  args: "info jsr/module_graph/main.ts",
-  output: "jsr/module_graph/main_info.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(deps_run {
-  args: "run jsr/deps/main.ts",
-  output: "jsr/deps/main.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(deps_info {
-  args: "info jsr/deps/main.ts",
-  output: "jsr/deps/main_info.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-});
-
-itest!(import_https_url_analyzable {
-  args: "run -A jsr/import_https_url/analyzable.ts",
-  output: "jsr/import_https_url/analyzable.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-  exit_code: 1,
-});
-
-itest!(import_https_url_unanalyzable {
-  args: "run -A jsr/import_https_url/unanalyzable.ts",
-  output: "jsr/import_https_url/unanalyzable.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-  exit_code: 1,
-});
-
-itest!(subset_type_graph {
-  args: "check --all jsr/subset_type_graph/main.ts",
-  output: "jsr/subset_type_graph/main.check.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-  exit_code: 1,
-});
-
 #[test]
 fn fast_check_cache() {
   let test_context = TestContextBuilder::for_jsr().use_temp_cwd().build();
@@ -197,14 +121,6 @@ export function asdf(a: number) { let err: number = ''; return Math.random(); }
   );
 }
 
-itest!(version_not_found {
-  args: "run jsr/version_not_found/main.ts",
-  output: "jsr/version_not_found/main.out",
-  envs: env_vars_for_jsr_tests(),
-  http_server: true,
-  exit_code: 1,
-});
-
 #[test]
 fn specifiers_in_lockfile() {
   let test_context = TestContextBuilder::for_jsr().use_temp_cwd().build();
@@ -478,19 +394,3 @@ fn set_lockfile_pkg_integrity(
     .unwrap()
     .integrity = integrity.to_string();
 }
-
-itest!(jsx_with_no_pragmas {
-  args: "run jsr/jsx_with_no_pragmas/main.ts",
-  output: "jsr/jsx_with_no_pragmas/main.out",
-  envs: env_vars_for_jsr_npm_tests(),
-  http_server: true,
-  exit_code: 1,
-});
-
-itest!(jsx_with_pragmas {
-  args: "run jsr/jsx_with_pragmas/main.ts",
-  output: "jsr/jsx_with_pragmas/main.out",
-  envs: env_vars_for_jsr_npm_tests(),
-  http_server: true,
-  exit_code: 0,
-});
diff --git a/tests/integration/jupyter_tests.rs b/tests/integration/jupyter_tests.rs
deleted file mode 100644
index 5478612585..0000000000
--- a/tests/integration/jupyter_tests.rs
+++ /dev/null
@@ -1,10 +0,0 @@
-// Copyright 2018-2024 the Deno authors. All rights reserved. MIT license.
-
-use test_util::itest;
-
-itest!(jupyter_install_command_not_exists {
-  args: "jupyter --install",
-  output: "jupyter/install_command_not_exists.out",
-  envs: vec![("PATH".to_string(), "".to_string())],
-  exit_code: 1,
-});
diff --git a/tests/integration/mod.rs b/tests/integration/mod.rs
index 9253cae32e..40dd880b54 100644
--- a/tests/integration/mod.rs
+++ b/tests/integration/mod.rs
@@ -38,8 +38,6 @@ mod install;
 mod js_unit_tests;
 #[path = "jsr_tests.rs"]
 mod jsr;
-#[path = "jupyter_tests.rs"]
-mod jupyter;
 #[path = "lint_tests.rs"]
 mod lint;
 #[path = "lsp_tests.rs"]
diff --git a/tests/specs/coverage/no_files_found/__test__.jsonc b/tests/specs/coverage/no_files_found/__test__.jsonc
new file mode 100644
index 0000000000..3637b41fe3
--- /dev/null
+++ b/tests/specs/coverage/no_files_found/__test__.jsonc
@@ -0,0 +1,5 @@
+{
+  "args": "coverage does_not_exit.js",
+  "output": "does_not_exist.out",
+  "exitCode": 1
+}
diff --git a/tests/testdata/coverage/doesnt_exist.out b/tests/specs/coverage/no_files_found/does_not_exist.out
similarity index 100%
rename from tests/testdata/coverage/doesnt_exist.out
rename to tests/specs/coverage/no_files_found/does_not_exist.out
diff --git a/tests/specs/flags/help_flag/__test__.jsonc b/tests/specs/flags/help_flag/__test__.jsonc
new file mode 100644
index 0000000000..4cf04538bc
--- /dev/null
+++ b/tests/specs/flags/help_flag/__test__.jsonc
@@ -0,0 +1,4 @@
+{
+  "args": "--help",
+  "output": "[WILDCARD]"
+}
diff --git a/tests/specs/flags/types/__test__.jsonc b/tests/specs/flags/types/__test__.jsonc
new file mode 100644
index 0000000000..ce0d61d348
--- /dev/null
+++ b/tests/specs/flags/types/__test__.jsonc
@@ -0,0 +1,4 @@
+{
+  "args": "types",
+  "output": "types.out"
+}
diff --git a/tests/testdata/types/types.out b/tests/specs/flags/types/types.out
similarity index 100%
rename from tests/testdata/types/types.out
rename to tests/specs/flags/types/types.out
diff --git a/tests/specs/flags/version_flag/__test__.jsonc b/tests/specs/flags/version_flag/__test__.jsonc
new file mode 100644
index 0000000000..e8c3c16468
--- /dev/null
+++ b/tests/specs/flags/version_flag/__test__.jsonc
@@ -0,0 +1,9 @@
+{
+  "steps": [{
+    "args": "-V",
+    "output": "[WILDCARD]"
+  }, {
+    "args": "--version",
+    "output": "[WILDCARD]"
+  }]
+}
diff --git a/tests/specs/jsr/deps/__test__.jsonc b/tests/specs/jsr/deps/__test__.jsonc
new file mode 100644
index 0000000000..2528003b51
--- /dev/null
+++ b/tests/specs/jsr/deps/__test__.jsonc
@@ -0,0 +1,11 @@
+{
+  "base": "jsr",
+  "steps": [{
+    "args": "run main.ts",
+    "output": "main.out"
+  }, {
+    "cleanDenoDir": true,
+    "args": "info main.ts",
+    "output": "main_info.out"
+  }]
+}
diff --git a/tests/testdata/jsr/deps/main.out b/tests/specs/jsr/deps/main.out
similarity index 100%
rename from tests/testdata/jsr/deps/main.out
rename to tests/specs/jsr/deps/main.out
diff --git a/tests/testdata/jsr/deps/main.ts b/tests/specs/jsr/deps/main.ts
similarity index 100%
rename from tests/testdata/jsr/deps/main.ts
rename to tests/specs/jsr/deps/main.ts
diff --git a/tests/testdata/jsr/deps/main_info.out b/tests/specs/jsr/deps/main_info.out
similarity index 100%
rename from tests/testdata/jsr/deps/main_info.out
rename to tests/specs/jsr/deps/main_info.out
diff --git a/tests/specs/jsr/import_https_url/__test__.jsonc b/tests/specs/jsr/import_https_url/__test__.jsonc
new file mode 100644
index 0000000000..597b8e009e
--- /dev/null
+++ b/tests/specs/jsr/import_https_url/__test__.jsonc
@@ -0,0 +1,13 @@
+{
+  "base": "jsr",
+  "steps": [{
+    "args": "run -A analyzable.ts",
+    "output": "analyzable.out",
+    "exitCode": 1
+  }, {
+    "cleanDenoDir": true,
+    "args": "run -A nonanalyzable.ts",
+    "output": "nonanalyzable.out",
+    "exitCode": 1
+  }]
+}
diff --git a/tests/testdata/jsr/import_https_url/analyzable.out b/tests/specs/jsr/import_https_url/analyzable.out
similarity index 100%
rename from tests/testdata/jsr/import_https_url/analyzable.out
rename to tests/specs/jsr/import_https_url/analyzable.out
diff --git a/tests/testdata/jsr/import_https_url/analyzable.ts b/tests/specs/jsr/import_https_url/analyzable.ts
similarity index 100%
rename from tests/testdata/jsr/import_https_url/analyzable.ts
rename to tests/specs/jsr/import_https_url/analyzable.ts
diff --git a/tests/testdata/jsr/import_https_url/unanalyzable.out b/tests/specs/jsr/import_https_url/nonanalyzable.out
similarity index 100%
rename from tests/testdata/jsr/import_https_url/unanalyzable.out
rename to tests/specs/jsr/import_https_url/nonanalyzable.out
diff --git a/tests/testdata/jsr/import_https_url/unanalyzable.ts b/tests/specs/jsr/import_https_url/nonanalyzable.ts
similarity index 100%
rename from tests/testdata/jsr/import_https_url/unanalyzable.ts
rename to tests/specs/jsr/import_https_url/nonanalyzable.ts
diff --git a/tests/specs/jsr/jsx_with_no_pragmas/__test__.jsonc b/tests/specs/jsr/jsx_with_no_pragmas/__test__.jsonc
new file mode 100644
index 0000000000..c2a269ea64
--- /dev/null
+++ b/tests/specs/jsr/jsx_with_no_pragmas/__test__.jsonc
@@ -0,0 +1,6 @@
+{
+  "base": "jsr",
+  "args": "run main.ts",
+  "output": "main.out",
+  "exitCode": 1
+}
diff --git a/tests/testdata/jsr/jsx_with_no_pragmas/main.out b/tests/specs/jsr/jsx_with_no_pragmas/main.out
similarity index 100%
rename from tests/testdata/jsr/jsx_with_no_pragmas/main.out
rename to tests/specs/jsr/jsx_with_no_pragmas/main.out
diff --git a/tests/testdata/jsr/jsx_with_no_pragmas/main.ts b/tests/specs/jsr/jsx_with_no_pragmas/main.ts
similarity index 100%
rename from tests/testdata/jsr/jsx_with_no_pragmas/main.ts
rename to tests/specs/jsr/jsx_with_no_pragmas/main.ts
diff --git a/tests/specs/jsr/jsx_with_pragmas/__test__.jsonc b/tests/specs/jsr/jsx_with_pragmas/__test__.jsonc
new file mode 100644
index 0000000000..a6fa5c7c3f
--- /dev/null
+++ b/tests/specs/jsr/jsx_with_pragmas/__test__.jsonc
@@ -0,0 +1,5 @@
+{
+  "base": "jsr",
+  "args": "run main.ts",
+  "output": "main.out"
+}
diff --git a/tests/testdata/jsr/jsx_with_pragmas/main.out b/tests/specs/jsr/jsx_with_pragmas/main.out
similarity index 100%
rename from tests/testdata/jsr/jsx_with_pragmas/main.out
rename to tests/specs/jsr/jsx_with_pragmas/main.out
diff --git a/tests/testdata/jsr/jsx_with_pragmas/main.ts b/tests/specs/jsr/jsx_with_pragmas/main.ts
similarity index 100%
rename from tests/testdata/jsr/jsx_with_pragmas/main.ts
rename to tests/specs/jsr/jsx_with_pragmas/main.ts
diff --git a/tests/specs/jsr/module_graph/__test__.jsonc b/tests/specs/jsr/module_graph/__test__.jsonc
new file mode 100644
index 0000000000..2528003b51
--- /dev/null
+++ b/tests/specs/jsr/module_graph/__test__.jsonc
@@ -0,0 +1,11 @@
+{
+  "base": "jsr",
+  "steps": [{
+    "args": "run main.ts",
+    "output": "main.out"
+  }, {
+    "cleanDenoDir": true,
+    "args": "info main.ts",
+    "output": "main_info.out"
+  }]
+}
diff --git a/tests/testdata/jsr/module_graph/main.out b/tests/specs/jsr/module_graph/main.out
similarity index 100%
rename from tests/testdata/jsr/module_graph/main.out
rename to tests/specs/jsr/module_graph/main.out
diff --git a/tests/testdata/jsr/module_graph/main.ts b/tests/specs/jsr/module_graph/main.ts
similarity index 100%
rename from tests/testdata/jsr/module_graph/main.ts
rename to tests/specs/jsr/module_graph/main.ts
diff --git a/tests/testdata/jsr/module_graph/main_info.out b/tests/specs/jsr/module_graph/main_info.out
similarity index 100%
rename from tests/testdata/jsr/module_graph/main_info.out
rename to tests/specs/jsr/module_graph/main_info.out
diff --git a/tests/specs/jsr/no_module_graph/__test__.jsonc b/tests/specs/jsr/no_module_graph/__test__.jsonc
new file mode 100644
index 0000000000..b3915ecef7
--- /dev/null
+++ b/tests/specs/jsr/no_module_graph/__test__.jsonc
@@ -0,0 +1,14 @@
+{
+  "base": "jsr",
+  "steps": [{
+    "args": "run main.ts",
+    "output": "main.out"
+  }, {
+    "cleanDenoDir": true,
+    "args": "info main.ts",
+    "output": "main_info.out"
+  }, {
+    "args": "run --quiet multiple.ts",
+    "output": "multiple.out"
+  }]
+}
diff --git a/tests/testdata/jsr/no_module_graph/main.out b/tests/specs/jsr/no_module_graph/main.out
similarity index 100%
rename from tests/testdata/jsr/no_module_graph/main.out
rename to tests/specs/jsr/no_module_graph/main.out
diff --git a/tests/testdata/jsr/no_module_graph/main.ts b/tests/specs/jsr/no_module_graph/main.ts
similarity index 100%
rename from tests/testdata/jsr/no_module_graph/main.ts
rename to tests/specs/jsr/no_module_graph/main.ts
diff --git a/tests/testdata/jsr/no_module_graph/main_info.out b/tests/specs/jsr/no_module_graph/main_info.out
similarity index 100%
rename from tests/testdata/jsr/no_module_graph/main_info.out
rename to tests/specs/jsr/no_module_graph/main_info.out
diff --git a/tests/testdata/jsr/no_module_graph/multiple.out b/tests/specs/jsr/no_module_graph/multiple.out
similarity index 100%
rename from tests/testdata/jsr/no_module_graph/multiple.out
rename to tests/specs/jsr/no_module_graph/multiple.out
diff --git a/tests/testdata/jsr/no_module_graph/multiple.ts b/tests/specs/jsr/no_module_graph/multiple.ts
similarity index 100%
rename from tests/testdata/jsr/no_module_graph/multiple.ts
rename to tests/specs/jsr/no_module_graph/multiple.ts
diff --git a/tests/specs/jsr/subset_type_graph/__test__.jsonc b/tests/specs/jsr/subset_type_graph/__test__.jsonc
new file mode 100644
index 0000000000..85d04ed54d
--- /dev/null
+++ b/tests/specs/jsr/subset_type_graph/__test__.jsonc
@@ -0,0 +1,6 @@
+{
+  "base": "jsr",
+  "args": "check --all main.ts",
+  "output": "main.check.out",
+  "exitCode": 1
+}
diff --git a/tests/testdata/jsr/subset_type_graph/main.check.out b/tests/specs/jsr/subset_type_graph/main.check.out
similarity index 100%
rename from tests/testdata/jsr/subset_type_graph/main.check.out
rename to tests/specs/jsr/subset_type_graph/main.check.out
diff --git a/tests/testdata/jsr/subset_type_graph/main.ts b/tests/specs/jsr/subset_type_graph/main.ts
similarity index 100%
rename from tests/testdata/jsr/subset_type_graph/main.ts
rename to tests/specs/jsr/subset_type_graph/main.ts
diff --git a/tests/specs/jsr/version_not_found/__test__.jsonc b/tests/specs/jsr/version_not_found/__test__.jsonc
new file mode 100644
index 0000000000..36335985fe
--- /dev/null
+++ b/tests/specs/jsr/version_not_found/__test__.jsonc
@@ -0,0 +1,6 @@
+{
+  "base": "jsr",
+  "args": "run -A main.ts",
+  "output": "main.out",
+  "exitCode": 1
+}
diff --git a/tests/testdata/jsr/version_not_found/main.out b/tests/specs/jsr/version_not_found/main.out
similarity index 100%
rename from tests/testdata/jsr/version_not_found/main.out
rename to tests/specs/jsr/version_not_found/main.out
diff --git a/tests/testdata/jsr/version_not_found/main.ts b/tests/specs/jsr/version_not_found/main.ts
similarity index 100%
rename from tests/testdata/jsr/version_not_found/main.ts
rename to tests/specs/jsr/version_not_found/main.ts
diff --git a/tests/specs/jupyter/install_command_not_exists/__test__.jsonc b/tests/specs/jupyter/install_command_not_exists/__test__.jsonc
new file mode 100644
index 0000000000..9552157bd4
--- /dev/null
+++ b/tests/specs/jupyter/install_command_not_exists/__test__.jsonc
@@ -0,0 +1,8 @@
+{
+  "args": "jupyter --install",
+  "output": "install_command_not_exists.out",
+  "envs": {
+    "PATH": ""
+  },
+  "exitCode": 1
+}
diff --git a/tests/testdata/jupyter/install_command_not_exists.out b/tests/specs/jupyter/install_command_not_exists/install_command_not_exists.out
similarity index 100%
rename from tests/testdata/jupyter/install_command_not_exists.out
rename to tests/specs/jupyter/install_command_not_exists/install_command_not_exists.out
diff --git a/tests/specs/mod.rs b/tests/specs/mod.rs
index d010de43de..feed098711 100644
--- a/tests/specs/mod.rs
+++ b/tests/specs/mod.rs
@@ -1,6 +1,7 @@
 // Copyright 2018-2024 the Deno authors. All rights reserved. MIT license.
 
 use std::cell::RefCell;
+use std::collections::HashMap;
 use std::collections::HashSet;
 use std::panic::AssertUnwindSafe;
 use std::rc::Rc;
@@ -127,15 +128,13 @@ fn run_test(test: &Test, diagnostic_logger: Rc<RefCell<Vec<u8>>>) {
       context.deno_dir().path().remove_dir_all();
     }
 
-    let test_output_path = cwd.join(&step.output);
-    if !test_output_path.to_string_lossy().ends_with(".out") {
-      panic!(
-        "Use the .out extension for output files (invalid: {})",
-        test_output_path
-      );
-    }
-    let expected_output = test_output_path.read_to_string();
-    let command = context.new_command();
+    let expected_output = if step.output.ends_with(".out") {
+      let test_output_path = cwd.join(&step.output);
+      test_output_path.read_to_string()
+    } else {
+      step.output.clone()
+    };
+    let command = context.new_command().envs(&step.envs);
     let command = match &step.args {
       VecOrString::Vec(args) => command.args_vec(args),
       VecOrString::String(text) => command.args(text),
@@ -195,6 +194,8 @@ struct StepMetaData {
   #[serde(default)]
   pub clean_deno_dir: bool,
   pub args: VecOrString,
+  #[serde(default)]
+  pub envs: HashMap<String, String>,
   pub output: String,
   #[serde(default)]
   pub exit_code: i32,
@@ -283,14 +284,6 @@ fn collect_tests() -> Vec<TestCategory> {
 
       let test_dir = PathRef::new(entry.path());
       let metadata_path = test_dir.join("__test__.jsonc");
-      if !metadata_path.is_file() {
-        let json_path = test_dir.join("__test__.json");
-        if json_path.is_file() {
-          // automatically rename to jsonc
-          json_path.rename(&metadata_path);
-        }
-      }
-
       let metadata_value = metadata_path.read_jsonc_value();
       // checking for "steps" leads to a more targeted error message
       // instead of when deserializing an untagged enum