diff --git a/.github/workflows/framework.yaml b/.github/workflows/framework_tools.yaml
similarity index 95%
rename from .github/workflows/framework.yaml
rename to .github/workflows/framework_tools.yaml
index dcc5e7614b15..a06f6e469aad 100644
--- a/.github/workflows/framework.yaml
+++ b/.github/workflows/framework_tools.yaml
@@ -1,4 +1,4 @@
-name: Framework tests
+name: Tool framework tests
 on:
   push:
     paths-ignore:
@@ -64,9 +64,9 @@ jobs:
         uses: actions/cache@v4
         with:
           path: 'galaxy root/.venv'
-          key: gxy-venv-${{ runner.os }}-${{ steps.full-python-version.outputs.version }}-${{ hashFiles('galaxy root/requirements.txt') }}-framework
+          key: gxy-venv-${{ runner.os }}-${{ steps.full-python-version.outputs.version }}-${{ hashFiles('galaxy root/requirements.txt') }}-framework-tools
       - name: Run tests
-        run: ./run_tests.sh --coverage --framework
+        run: ./run_tests.sh --coverage --framework-tools
         working-directory: 'galaxy root'
       - uses: codecov/codecov-action@v3
         with:
diff --git a/.github/workflows/framework_workflows.yaml b/.github/workflows/framework_workflows.yaml
new file mode 100644
index 000000000000..accbbd4c3736
--- /dev/null
+++ b/.github/workflows/framework_workflows.yaml
@@ -0,0 +1,79 @@
+name: Workflow framework tests
+on:
+  push:
+    paths-ignore:
+      - 'client/**'
+      - 'doc/**'
+      - 'lib/galaxy_test/selenium/**'
+  pull_request:
+    paths-ignore:
+      - 'client/**'
+      - 'doc/**'
+      - 'lib/galaxy_test/selenium/**'
+  schedule:
+    # Run at midnight UTC every Tuesday
+    - cron: '0 0 * * 2'
+env:
+  GALAXY_TEST_DBURI: 'postgresql://postgres:postgres@localhost:5432/galaxy?client_encoding=utf8'
+  GALAXY_TEST_RAISE_EXCEPTION_ON_HISTORYLESS_HDA: '1'
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+jobs:
+  test:
+    name: Test
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ['3.8']
+    services:
+      postgres:
+        image: postgres:13
+        env:
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: postgres
+        ports:
+          - 5432:5432
+    steps:
+      - if: github.event_name == 'schedule'
+        run: |
+          echo "GALAXY_CONFIG_OVERRIDE_METADATA_STRATEGY=extended" >> $GITHUB_ENV
+          echo "GALAXY_CONFIG_OVERRIDE_OUTPUTS_TO_WORKING_DIRECTORY=true" >> $GITHUB_ENV
+      - uses: actions/checkout@v4
+        with:
+          path: 'galaxy root'
+      - uses: actions/setup-node@v4
+        with:
+          node-version: '18.12.1'
+          cache: 'yarn'
+          cache-dependency-path: 'galaxy root/client/yarn.lock'
+      - uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Get full Python version
+        id: full-python-version
+        shell: bash
+        run: echo "version=$(python -c 'import sys; print("-".join(str(v) for v in sys.version_info))')" >> $GITHUB_OUTPUT
+      - name: Cache pip dir
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pip
+          key: pip-cache-${{ matrix.python-version }}-${{ hashFiles('galaxy root/requirements.txt') }}
+      - name: Cache galaxy venv
+        uses: actions/cache@v4
+        with:
+          path: 'galaxy root/.venv'
+          key: gxy-venv-${{ runner.os }}-${{ steps.full-python-version.outputs.version }}-${{ hashFiles('galaxy root/requirements.txt') }}-framework-workflows
+      - name: Run tests
+        run: ./run_tests.sh --coverage --framework-workflows
+        working-directory: 'galaxy root'
+      - uses: codecov/codecov-action@v3
+        with:
+          flags: framework
+          working-directory: 'galaxy root'
+      - uses: actions/upload-artifact@v4
+        if: failure()
+        with:
+          name: Framework test results (${{ matrix.python-version }})
+          path: 'galaxy root/run_framework_workflows_tests.html'
diff --git a/lib/galaxy/tool_util/parser/yaml.py b/lib/galaxy/tool_util/parser/yaml.py
index c354bfd33d49..6813db0f5211 100644
--- a/lib/galaxy/tool_util/parser/yaml.py
+++ b/lib/galaxy/tool_util/parser/yaml.py
@@ -310,6 +310,8 @@ def expand_dict_form(item):
     return assert_list or None  # XML variant is None if no assertions made
 
 
+# Planemo depends on this and was never updated unfortunately.
+# https://github.com/galaxyproject/planemo/blob/master/planemo/test/_check_output.py
 __to_test_assert_list = to_test_assert_list
 
 
diff --git a/lib/galaxy/tool_util/verify/__init__.py b/lib/galaxy/tool_util/verify/__init__.py
index ec84f23ae345..2bcb31106bfb 100644
--- a/lib/galaxy/tool_util/verify/__init__.py
+++ b/lib/galaxy/tool_util/verify/__init__.py
@@ -45,6 +45,7 @@
     DEFAULT_METRIC,
     DEFAULT_PIN_LABELS,
 )
+from galaxy.tool_util.parser.yaml import to_test_assert_list
 from galaxy.util import unicodify
 from galaxy.util.compression_utils import get_fileobj
 from .asserts import verify_assertions
@@ -56,6 +57,8 @@
 log = logging.getLogger(__name__)
 
 DEFAULT_TEST_DATA_RESOLVER = TestDataResolver()
+GetFilenameT = Optional[Callable[[str], str]]
+GetLocationT = Optional[Callable[[str], str]]
 
 
 def verify(
@@ -64,7 +67,7 @@ def verify(
     attributes: Optional[Dict[str, Any]],
     filename: Optional[str] = None,
     get_filecontent: Optional[Callable[[str], bytes]] = None,
-    get_filename: Optional[Callable[[str], str]] = None,
+    get_filename: GetFilenameT = None,
     keep_outputs_dir: Optional[str] = None,
     verify_extra_files: Optional[Callable] = None,
     mode="file",
@@ -585,3 +588,56 @@ def files_image_diff(file1: str, file2: str, attributes: Optional[Dict[str, Any]
     distance_eps = attributes.get("eps", DEFAULT_EPS)
     if distance > distance_eps:
         raise AssertionError(f"Image difference {distance} exceeds eps={distance_eps}.")
+
+
+# TODO: After tool-util with this included is published, fefactor planemo.test._check_output
+# to use this function. There is already a comment there about breaking fewer abstractions.
+# https://github.com/galaxyproject/planemo/blob/master/planemo/test/_check_output.py
+def verify_file_path_against_dict(
+    get_filename: GetFilenameT,
+    get_location: GetLocationT,
+    path: str,
+    output_content: bytes,
+    test_properties,
+    test_data_target_dir: Optional[str] = None,
+) -> None:
+    with open(path, "rb") as f:
+        output_content = f.read()
+    item_label = f"Output with path {path}"
+    verify_file_contents_against_dict(
+        get_filename, get_location, item_label, output_content, test_properties, test_data_target_dir
+    )
+
+
+def verify_file_contents_against_dict(
+    get_filename: GetFilenameT,
+    get_location: GetLocationT,
+    item_label: str,
+    output_content: bytes,
+    test_properties,
+    test_data_target_dir: Optional[str] = None,
+) -> None:
+    # Support Galaxy-like file location (using "file") or CWL-like ("path" or "location").
+    expected_file = test_properties.get("file", None)
+    if expected_file is None:
+        expected_file = test_properties.get("path", None)
+    if expected_file is None:
+        location = test_properties.get("location")
+        if location:
+            if location.startswith(("http://", "https://")):
+                assert get_location
+                expected_file = get_location(location)
+            else:
+                expected_file = location.split("file://", 1)[-1]
+
+    if "asserts" in test_properties:
+        test_properties["assert_list"] = to_test_assert_list(test_properties["asserts"])
+    verify(
+        item_label,
+        output_content,
+        attributes=test_properties,
+        filename=expected_file,
+        get_filename=get_filename,
+        keep_outputs_dir=test_data_target_dir,
+        verify_extra_files=None,
+    )
diff --git a/lib/galaxy/tool_util/verify/interactor.py b/lib/galaxy/tool_util/verify/interactor.py
index 77a27609a3ad..e403d08fa1aa 100644
--- a/lib/galaxy/tool_util/verify/interactor.py
+++ b/lib/galaxy/tool_util/verify/interactor.py
@@ -346,18 +346,7 @@ def _verify_metadata(self, history_id, hid, attributes):
         `dbkey` and `tags` all map to the API description directly. Other metadata attributes
         are assumed to be datatype-specific and mapped with a prefix of `metadata_`.
         """
-        metadata = attributes.get("metadata", {}).copy()
-        for key in metadata.copy().keys():
-            if key not in ["name", "info", "tags", "created_from_basename"]:
-                new_key = f"metadata_{key}"
-                metadata[new_key] = metadata[key]
-                del metadata[key]
-            elif key == "info":
-                metadata["misc_info"] = metadata["info"]
-                del metadata["info"]
-        expected_file_type = attributes.get("ftype", None)
-        if expected_file_type:
-            metadata["file_ext"] = expected_file_type
+        metadata = get_metadata_to_test(attributes)
 
         if metadata:
 
@@ -370,29 +359,7 @@ def wait_for_content():
                     return None
 
             dataset = wait_on(wait_for_content, desc="dataset metadata", timeout=10)
-
-            for key, value in metadata.items():
-                try:
-                    dataset_value = dataset.get(key, None)
-
-                    def compare(val, expected):
-                        if str(val) != str(expected):
-                            raise Exception(
-                                f"Dataset metadata verification for [{key}] failed, expected [{value}] but found [{dataset_value}]. Dataset API value was [{dataset}]."  # noqa: B023
-                            )
-
-                    if isinstance(dataset_value, list):
-                        value = str(value).split(",")
-                        if len(value) != len(dataset_value):
-                            raise Exception(
-                                f"Dataset metadata verification for [{key}] failed, expected [{value}] but found [{dataset_value}], lists differ in length. Dataset API value was [{dataset}]."
-                            )
-                        for val, expected in zip(dataset_value, value):
-                            compare(val, expected)
-                    else:
-                        compare(dataset_value, value)
-                except KeyError:
-                    raise Exception(f"Failed to verify dataset metadata, metadata key [{key}] was not found.")
+            compare_expected_metadata_to_api_response(metadata, dataset)
 
     def wait_for_job(self, job_id: str, history_id: Optional[str] = None, maxseconds=DEFAULT_TOOL_TEST_WAIT) -> None:
         self.wait_for(lambda: self.__job_ready(job_id, history_id), maxseconds=maxseconds)
@@ -1934,3 +1901,45 @@ def test_data_iter(required_files):
                 raise Exception(f"edit_attributes type ({edit_att.get('type', None)}) is unimplemented")
 
         yield data_dict
+
+
+def compare_expected_metadata_to_api_response(metadata: dict, dataset: dict) -> None:
+    for key, value in metadata.items():
+        try:
+            dataset_value = dataset.get(key, None)
+
+            def compare(val, expected):
+                if str(val) != str(expected):
+                    raise Exception(
+                        f"Dataset metadata verification for [{key}] failed, expected [{value}] but found [{dataset_value}]. Dataset API value was [{dataset}]."  # noqa: B023
+                    )
+
+            if isinstance(dataset_value, list):
+                value = str(value).split(",")
+                if len(value) != len(dataset_value):
+                    raise Exception(
+                        f"Dataset metadata verification for [{key}] failed, expected [{value}] but found [{dataset_value}], lists differ in length. Dataset API value was [{dataset}]."
+                    )
+                for val, expected in zip(dataset_value, value):
+                    compare(val, expected)
+            else:
+                compare(dataset_value, value)
+        except KeyError:
+            raise Exception(f"Failed to verify dataset metadata, metadata key [{key}] was not found.")
+
+
+def get_metadata_to_test(test_properties: dict) -> dict:
+    """Fetch out metadata to test from test property dict and adapt it to keys the API produces."""
+    metadata = test_properties.get("metadata", {}).copy()
+    for key in metadata.copy().keys():
+        if key not in ["name", "info", "tags", "created_from_basename"]:
+            new_key = f"metadata_{key}"
+            metadata[new_key] = metadata[key]
+            del metadata[key]
+        elif key == "info":
+            metadata["misc_info"] = metadata["info"]
+            del metadata["info"]
+    expected_file_type = test_properties.get("ftype", None)
+    if expected_file_type:
+        metadata["file_ext"] = expected_file_type
+    return metadata
diff --git a/lib/galaxy_test/api/test_workflows.py b/lib/galaxy_test/api/test_workflows.py
index c42da227984e..88d4112966fd 100644
--- a/lib/galaxy_test/api/test_workflows.py
+++ b/lib/galaxy_test/api/test_workflows.py
@@ -1495,36 +1495,6 @@ def test_run_workflow_by_name(self):
     def test_run_workflow(self):
         self.__run_cat_workflow(inputs_by="step_id")
 
-    @skip_without_tool("multiple_versions")
-    def test_run_versioned_tools(self):
-        with self.dataset_populator.test_history() as history_01_id:
-            workflow_version_01 = self._upload_yaml_workflow(
-                """
-class: GalaxyWorkflow
-steps:
-  multiple:
-    tool_id: multiple_versions
-    tool_version: "0.1"
-    state:
-      inttest: 0
-"""
-            )
-            self.workflow_populator.invoke_workflow_and_wait(workflow_version_01, history_id=history_01_id)
-
-        with self.dataset_populator.test_history() as history_02_id:
-            workflow_version_02 = self._upload_yaml_workflow(
-                """
-class: GalaxyWorkflow
-steps:
-  multiple:
-    tool_id: multiple_versions
-    tool_version: "0.2"
-    state:
-      inttest: 1
-"""
-            )
-            self.workflow_populator.invoke_workflow_and_wait(workflow_version_02, history_id=history_02_id)
-
     def __run_cat_workflow(self, inputs_by):
         workflow = self.workflow_populator.load_workflow(name="test_for_run")
         workflow["steps"]["0"]["uuid"] = str(uuid4())
@@ -2771,93 +2741,6 @@ def test_run_subworkflow_replacement_parameters(self):
             details = self.dataset_populator.get_history_dataset_details(history_id)
             assert details["name"] == "moocow suffix"
 
-    @skip_without_tool("create_2")
-    def test_placements_from_text_inputs(self):
-        with self.dataset_populator.test_history() as history_id:
-            run_def = """
-class: GalaxyWorkflow
-inputs: []
-steps:
-  create_2:
-    tool_id: create_2
-    state:
-      sleep_time: 0
-    outputs:
-      out_file1:
-        rename: "${replaceme} name"
-      out_file2:
-        rename: "${replaceme} name 2"
-test_data:
-  replacement_parameters:
-    replaceme: moocow
-"""
-
-            self._run_jobs(run_def, history_id=history_id)
-            details = self.dataset_populator.get_history_dataset_details(history_id)
-            assert details["name"] == "moocow name 2"
-
-            run_def = """
-class: GalaxyWorkflow
-inputs:
-  replaceme: text
-steps:
-  create_2:
-    tool_id: create_2
-    state:
-      sleep_time: 0
-    outputs:
-      out_file1:
-        rename: "${replaceme} name"
-      out_file2:
-        rename: "${replaceme} name 2"
-test_data:
-  replaceme:
-    value: moocow
-    type: raw
-"""
-            self._run_jobs(run_def, history_id=history_id)
-            details = self.dataset_populator.get_history_dataset_details(history_id)
-            assert details["name"] == "moocow name 2", details["name"]
-
-    def test_placements_from_text_inputs_nested(self):
-        with self.dataset_populator.test_history() as history_id:
-            run_def = """
-class: GalaxyWorkflow
-inputs:
-  replacemeouter: text
-steps:
-  nested_workflow:
-    run:
-      class: GalaxyWorkflow
-      inputs:
-        replacemeinner: text
-      outputs:
-        workflow_output_1:
-          outputSource: create_2/out_file1
-        workflow_output_2:
-          outputSource: create_2/out_file2
-      steps:
-        create_2:
-          tool_id: create_2
-          state:
-            sleep_time: 0
-          outputs:
-            out_file1:
-              rename: "${replacemeinner} name"
-            out_file2:
-              rename: "${replacemeinner} name 2"
-    in:
-      replacemeinner: replacemeouter
-
-test_data:
-  replacemeouter:
-    value: moocow
-    type: raw
-"""
-            self._run_jobs(run_def, history_id=history_id)
-            details = self.dataset_populator.get_history_dataset_details(history_id)
-            assert details["name"] == "moocow name 2", details["name"]
-
     @skip_without_tool("random_lines1")
     def test_run_runtime_parameters_after_pause(self):
         with self.dataset_populator.test_history() as history_id:
@@ -2917,111 +2800,6 @@ def run_test(workflow_text):
 
         run_test(NESTED_WORKFLOW_AUTO_LABELS_MODERN_SYNTAX)
 
-    @skip_without_tool("cat1")
-    @skip_without_tool("collection_paired_test")
-    def test_workflow_run_zip_collections(self):
-        with self.dataset_populator.test_history() as history_id:
-            workflow_id = self._upload_yaml_workflow(
-                """
-class: GalaxyWorkflow
-inputs:
-  test_input_1: data
-  test_input_2: data
-steps:
-  first_cat:
-    tool_id: cat1
-    in:
-      input1: test_input_1
-  zip_it:
-    tool_id: "__ZIP_COLLECTION__"
-    in:
-      input_forward: first_cat/out_file1
-      input_reverse: test_input_2
-  concat_pair:
-    tool_id: collection_paired_test
-    in:
-      f1: zip_it/output
-"""
-            )
-            hda1 = self.dataset_populator.new_dataset(history_id, content="samp1\t10.0\nsamp2\t20.0\n")
-            hda2 = self.dataset_populator.new_dataset(history_id, content="samp1\t20.0\nsamp2\t40.0\n")
-            self.dataset_populator.wait_for_history(history_id, assert_ok=True)
-            inputs = {
-                "0": self._ds_entry(hda1),
-                "1": self._ds_entry(hda2),
-            }
-            invocation_id = self.__invoke_workflow(workflow_id, inputs=inputs, history_id=history_id)
-            self.workflow_populator.wait_for_invocation_and_jobs(history_id, workflow_id, invocation_id)
-            content = self.dataset_populator.get_history_dataset_content(history_id)
-            assert content.strip() == "samp1\t10.0\nsamp2\t20.0\nsamp1\t20.0\nsamp2\t40.0"
-
-    @skip_without_tool("collection_paired_test")
-    def test_workflow_flatten(self):
-        with self.dataset_populator.test_history() as history_id:
-            self._run_jobs(
-                """
-class: GalaxyWorkflow
-steps:
-  nested:
-    tool_id: collection_creates_dynamic_nested
-    state:
-      sleep_time: 0
-      foo: 'dummy'
-  flatten:
-    tool_id: '__FLATTEN__'
-    state:
-      input:
-        $link: nested/list_output
-      join_identifier: '-'
-""",
-                test_data={},
-                history_id=history_id,
-            )
-            details = self.dataset_populator.get_history_collection_details(history_id, hid=14)
-            assert details["collection_type"] == "list"
-            elements = details["elements"]
-            identifiers = [e["element_identifier"] for e in elements]
-            assert len(identifiers) == 6
-            assert "oe1-ie1" in identifiers
-
-    @skip_without_tool("collection_paired_test")
-    def test_workflow_flatten_with_mapped_over_execution(self):
-        with self.dataset_populator.test_history() as history_id:
-            self._run_jobs(
-                r"""
-class: GalaxyWorkflow
-inputs:
-  input_fastqs: collection
-steps:
-  split_up:
-    tool_id: collection_split_on_column
-    in:
-      input1: input_fastqs
-  flatten:
-    tool_id: '__FLATTEN__'
-    in:
-      input: split_up/split_output
-    join_identifier: '-'
-test_data:
-  input_fastqs:
-    collection_type: list
-    elements:
-      - identifier: samp1
-        content: "0\n1"
-""",
-                history_id=history_id,
-            )
-            history = self._get(f"histories/{history_id}/contents").json()
-            flattened_collection = history[-1]
-            assert flattened_collection["history_content_type"] == "dataset_collection"
-            assert flattened_collection["collection_type"] == "list"
-            assert flattened_collection["element_count"] == 2
-            nested_collection = self.dataset_populator.get_history_collection_details(history_id, hid=3)
-            assert nested_collection["collection_type"] == "list:list"
-            assert nested_collection["element_count"] == 1
-            assert nested_collection["elements"][0]["object"]["populated"]
-            assert nested_collection["elements"][0]["object"]["element_count"] == 2
-
     @skip_without_tool("cat")
     def test_workflow_invocation_report_1(self):
         test_data = """
@@ -6249,52 +6027,6 @@ def test_run_rename_based_on_input_conditional_legacy_pja_reference(self):
             name = content["name"]
             assert name == "fastq1 suffix", name
 
-    @skip_without_tool("mapper2")
-    def test_run_rename_based_on_input_collection(self):
-        with self.dataset_populator.test_history() as history_id:
-            self._run_jobs(
-                """
-class: GalaxyWorkflow
-inputs:
-  fasta_input: data
-  fastq_inputs: data
-steps:
-  mapping:
-    tool_id: mapper2
-    state:
-      fastq_input:
-        fastq_input_selector: paired_collection
-        fastq_input1:
-          $link: fastq_inputs
-      reference:
-        $link: fasta_input
-    outputs:
-      out_file1:
-        rename: "#{fastq_input.fastq_input1 | basename} suffix"
-""",
-                test_data="""
-fasta_input:
-  value: 1.fasta
-  type: File
-  name: fasta1
-  file_type: fasta
-fastq_inputs:
-  collection_type: list
-  name: the_dataset_pair
-  elements:
-    - identifier: forward
-      value: 1.fastq
-      type: File
-    - identifier: reverse
-      value: 1.fastq
-      type: File
-""",
-                history_id=history_id,
-            )
-            content = self.dataset_populator.get_history_dataset_details(history_id, wait=True, assert_ok=True)
-            name = content["name"]
-            assert name == "the_dataset_pair suffix", name
-
     @skip_without_tool("collection_creates_pair")
     def test_run_hide_on_collection_output(self):
         with self.dataset_populator.test_history() as history_id:
@@ -7184,44 +6916,6 @@ def test_defaults_editor(self):
         put_response = self._update_workflow(workflow_id, workflow_object)
         assert put_response.status_code == 200
 
-    def test_empty_collection_sort(self, history_id):
-        self._run_workflow(
-            """class: GalaxyWorkflow
-inputs:
-  input: collection
-  filter_file: data
-steps:
-  filter_collection:
-    tool_id: __FILTER_FROM_FILE__
-    in:
-       input: input
-       how|filter_source: filter_file
-  sort_collection_1:
-    tool_id: __SORTLIST__
-    in:
-      input: filter_collection/output_filtered
-  sort_collection_2:
-    tool_id: __SORTLIST__
-    in:
-      input: filter_collection/output_discarded
-  merge_collection:
-    tool_id: __MERGE_COLLECTION__
-    in:
-      inputs_0|input: sort_collection_1/output
-      inputs_1|input: sort_collection_2/output
-test_data:
-  input:
-    collection_type: list
-    elements:
-      - identifier: i1
-        content: "0"
-  filter_file: i1
-""",
-            history_id=history_id,
-            wait=True,
-            assert_ok=True,
-        )
-
     @skip_without_tool("random_lines1")
     def test_run_replace_params_over_default_delayed(self):
         with self.dataset_populator.test_history() as history_id:
diff --git a/lib/galaxy_test/driver/driver_util.py b/lib/galaxy_test/driver/driver_util.py
index 77ab88ca0682..a2ee29f5f774 100644
--- a/lib/galaxy_test/driver/driver_util.py
+++ b/lib/galaxy_test/driver/driver_util.py
@@ -850,7 +850,8 @@ def _configure(self, config_object=None) -> None:
 
         default_tool_conf: Optional[str]
         datatypes_conf_override: Optional[str]
-        if getattr(config_object, "framework_tool_and_types", False):
+        framework_tools_and_types = getattr(config_object, "framework_tool_and_types", False)
+        if framework_tools_and_types:
             default_tool_conf = FRAMEWORK_SAMPLE_TOOLS_CONF
             datatypes_conf_override = FRAMEWORK_DATATYPES_CONF
         else:
diff --git a/lib/galaxy_test/workflow/__init__.py b/lib/galaxy_test/workflow/__init__.py
new file mode 100644
index 000000000000..2355ab50ab9a
--- /dev/null
+++ b/lib/galaxy_test/workflow/__init__.py
@@ -0,0 +1,7 @@
+"""Framework tests for checking workflow correctness and defining expected workflow behavior.
+
+This is meant to grow into the workflow based mirror of what the framework tests are for tools.
+api/test_workflows.py is still the place to test exceptional conditions, errors, etc... but tests
+of normal operation where semantics can be verified with simple inputs and outputs can now be
+placed in here.
+"""
diff --git a/lib/galaxy_test/workflow/conftest.py b/lib/galaxy_test/workflow/conftest.py
new file mode 100644
index 000000000000..1075bb4a374d
--- /dev/null
+++ b/lib/galaxy_test/workflow/conftest.py
@@ -0,0 +1,22 @@
+import os
+
+import pytest
+
+from galaxy_test.driver.driver_util import GalaxyTestDriver
+
+
+class ConfigObject:
+    framework_tool_and_types = True
+
+
+@pytest.fixture(scope="session")
+def real_driver():
+    if not os.environ.get("GALAXY_TEST_ENVIRONMENT_CONFIGURED"):
+        driver = GalaxyTestDriver()
+        driver.setup(ConfigObject)
+        try:
+            yield driver
+        finally:
+            driver.tear_down()
+    else:
+        yield None
diff --git a/lib/galaxy_test/workflow/empty_collection_sort.gxwf-tests.yml b/lib/galaxy_test/workflow/empty_collection_sort.gxwf-tests.yml
new file mode 100644
index 000000000000..ad83bb676b67
--- /dev/null
+++ b/lib/galaxy_test/workflow/empty_collection_sort.gxwf-tests.yml
@@ -0,0 +1,16 @@
+- doc: | 
+    Test to verify collection operations like the sort tool work fine with empty collections.
+  job:
+    input:
+      collection_type: list
+      elements:
+      - identifier: i1
+        content: "0"
+    filter_file: i1
+  outputs:
+    output:
+      elements:
+        i1:
+          asserts:
+          - that: has_text
+            text: "0"
diff --git a/lib/galaxy_test/workflow/empty_collection_sort.gxwf.yml b/lib/galaxy_test/workflow/empty_collection_sort.gxwf.yml
new file mode 100644
index 000000000000..7520bcd5d68f
--- /dev/null
+++ b/lib/galaxy_test/workflow/empty_collection_sort.gxwf.yml
@@ -0,0 +1,26 @@
+class: GalaxyWorkflow
+inputs:
+  input: collection
+  filter_file: data
+outputs:
+  output:
+    outputSource: merge_collection/output
+steps:
+  filter_collection:
+    tool_id: __FILTER_FROM_FILE__
+    in:
+       input: input
+       how|filter_source: filter_file
+  sort_collection_1:
+    tool_id: __SORTLIST__
+    in:
+      input: filter_collection/output_filtered
+  sort_collection_2:
+    tool_id: __SORTLIST__
+    in:
+      input: filter_collection/output_discarded
+  merge_collection:
+    tool_id: __MERGE_COLLECTION__
+    in:
+      inputs_0|input: sort_collection_1/output
+      inputs_1|input: sort_collection_2/output
diff --git a/lib/galaxy_test/workflow/flatten_collection.gxwf-tests.yml b/lib/galaxy_test/workflow/flatten_collection.gxwf-tests.yml
new file mode 100644
index 000000000000..1a2b5c65b596
--- /dev/null
+++ b/lib/galaxy_test/workflow/flatten_collection.gxwf-tests.yml
@@ -0,0 +1,18 @@
+- doc: | 
+    Test to verify collection flatten collection operation in the context of a workflow.
+  job: {}
+  outputs:
+    out:
+      elements:
+        'oe1-ie1':
+          asserts:
+          - that: has_text
+            text: "A"
+        'oe1-ie2':
+          asserts:
+          - that: has_text
+            text: "B"
+        'oe2-ie1':
+          asserts:
+          - that: has_text
+            text: "C"
diff --git a/lib/galaxy_test/workflow/flatten_collection.gxwf.yml b/lib/galaxy_test/workflow/flatten_collection.gxwf.yml
new file mode 100644
index 000000000000..993bd9dac005
--- /dev/null
+++ b/lib/galaxy_test/workflow/flatten_collection.gxwf.yml
@@ -0,0 +1,17 @@
+class: GalaxyWorkflow
+inputs: {}
+outputs:
+  out:
+    outputSource: flatten/output
+steps:
+  nested:
+    tool_id: collection_creates_dynamic_nested
+    state:
+      sleep_time: 0
+      foo: 'dummy'
+  flatten:
+    tool_id: '__FLATTEN__'
+    state:
+      join_identifier: '-'
+    in:
+      input: nested/list_output
diff --git a/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf-tests.yml b/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf-tests.yml
new file mode 100644
index 000000000000..5c54212815e9
--- /dev/null
+++ b/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf-tests.yml
@@ -0,0 +1,19 @@
+- doc: | 
+    Test to verify collection flatten collection operation mid workflow.
+  job:
+    input_fastqs:
+      collection_type: list
+      elements:
+        - identifier: samp1
+          content: "0 mycoolline\n1 mysecondline\n"
+  outputs:
+    out:
+      elements:
+        'samp1-0':
+          asserts:
+          - that: has_text
+            text: "mycoolline"
+        'samp1-1':
+          asserts:
+          - that: has_text
+            text: "mysecondline"
diff --git a/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf.yml b/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf.yml
new file mode 100644
index 000000000000..172eaa7d4b26
--- /dev/null
+++ b/lib/galaxy_test/workflow/flatten_collection_over_execution.gxwf.yml
@@ -0,0 +1,17 @@
+class: GalaxyWorkflow
+inputs:
+  input_fastqs: collection
+outputs:
+  out:
+    outputSource: flatten/output
+steps:
+  split_up:
+    tool_id: collection_split_on_column
+    in:
+      input1: input_fastqs
+  flatten:
+    tool_id: '__FLATTEN__'
+    state:
+      join_identifier: '-'
+    in:
+      input: split_up/split_output
diff --git a/lib/galaxy_test/workflow/multi_select_mapping.gxwf-tests.yml b/lib/galaxy_test/workflow/multi_select_mapping.gxwf-tests.yml
new file mode 100644
index 000000000000..065849a34082
--- /dev/null
+++ b/lib/galaxy_test/workflow/multi_select_mapping.gxwf-tests.yml
@@ -0,0 +1,27 @@
+- doc: |
+    Test to verify that mapping a list of a multi-select parameter maps over the job.
+    Ideally there would be a way to specify that the list should be reduced into the
+    parameter. It might work to have an expression tool to combine the list into a
+    string.
+  job:
+    input:
+      type: collection
+      collection_type: list
+      elements:
+        - identifier: the_example_2
+          content: '"ex2"'
+          ext: 'expression.json'
+        - identifier: the_example_5
+          content: '"ex5"'
+          ext: 'expression.json'
+  outputs:
+    output:
+      elements:
+        the_example_2:
+          asserts:
+          - that: has_text
+            text: 'ex2'
+        the_example_5:
+          asserts:
+          - that: has_text
+            text: 'ex5'
diff --git a/lib/galaxy_test/workflow/multi_select_mapping.gxwf.yml b/lib/galaxy_test/workflow/multi_select_mapping.gxwf.yml
new file mode 100644
index 000000000000..f810d022fb1a
--- /dev/null
+++ b/lib/galaxy_test/workflow/multi_select_mapping.gxwf.yml
@@ -0,0 +1,14 @@
+class: GalaxyWorkflow
+inputs:
+  input:
+    type: collection
+    collection_type: list
+outputs:
+  output:
+    outputSource: map_over_multi_select/output2
+steps:
+  map_over_multi_select:
+    tool_id: multi_select
+    in:
+      select_ex: input
+      select_optional: input
diff --git a/lib/galaxy_test/workflow/multiple_versions.gxwf-tests.yml b/lib/galaxy_test/workflow/multiple_versions.gxwf-tests.yml
new file mode 100644
index 000000000000..1d1db5dba980
--- /dev/null
+++ b/lib/galaxy_test/workflow/multiple_versions.gxwf-tests.yml
@@ -0,0 +1,19 @@
+- doc: | 
+    Test that workflows can run with multiple versions of tools and the version specified matches the one run.
+  job:
+    input:
+      value: 1
+      type: raw
+  outputs:
+    output_1:
+      asserts:
+        - that: has_text
+          text: 'Version 0.1'
+        - that: not_has_text
+          text: 'Version 0.2'
+    output_2:
+      asserts:
+        - that: has_text
+          text: 'Version 0.2'
+        - that: not_has_text
+          text: 'Version 0.1'
diff --git a/lib/galaxy_test/workflow/multiple_versions.gxwf.yml b/lib/galaxy_test/workflow/multiple_versions.gxwf.yml
new file mode 100644
index 000000000000..abcdd44e2ca8
--- /dev/null
+++ b/lib/galaxy_test/workflow/multiple_versions.gxwf.yml
@@ -0,0 +1,20 @@
+class: GalaxyWorkflow
+inputs:
+  input:
+    type: int
+outputs:
+  output_1:
+    outputSource: multiple_1/out_file1
+  output_2:
+    outputSource: multiple_2/out_file1
+steps:
+  multiple_1:
+    tool_id: multiple_versions
+    tool_version: "0.1"
+    in:
+      inttest: input
+  multiple_2:
+    tool_id: multiple_versions
+    tool_version: "0.2"
+    in:
+      inttest: input
diff --git a/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf-tests.yml b/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf-tests.yml
new file mode 100644
index 000000000000..a6b357635316
--- /dev/null
+++ b/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf-tests.yml
@@ -0,0 +1,22 @@
+- doc: | 
+    Test output dataset renaming when the target basename is based on an input collection.
+  job:
+    fasta_input:
+      value: 1.fasta
+      type: File
+      name: fasta1
+      file_type: fasta
+    fastq_inputs:
+      collection_type: list
+      name: the_dataset_pair
+      elements:
+        - identifier: forward
+          value: 1.fastq
+          type: File
+        - identifier: reverse
+          value: 1.fastq
+          type: File
+  outputs:
+    output:
+      metadata:
+        name: 'the_dataset_pair suffix'
diff --git a/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf.yml b/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf.yml
new file mode 100644
index 000000000000..0d6a19abfdd3
--- /dev/null
+++ b/lib/galaxy_test/workflow/rename_based_on_input_collection.gxwf.yml
@@ -0,0 +1,20 @@
+class: GalaxyWorkflow
+inputs:
+  fasta_input: data
+  fastq_inputs: data
+outputs:
+  output:
+    outputSource: mapping/out_file1
+steps:
+  mapping:
+    tool_id: mapper2
+    state:
+      fastq_input:
+        fastq_input_selector: paired_collection
+        fastq_input1:
+          $link: fastq_inputs
+      reference:
+        $link: fasta_input
+    outputs:
+      out_file1:
+        rename: "#{fastq_input.fastq_input1 | basename} suffix"
diff --git a/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf-tests.yml b/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf-tests.yml
new file mode 100644
index 000000000000..a3120f277072
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf-tests.yml
@@ -0,0 +1,12 @@
+- doc: |
+    Test to verify legacy behavior of implicit runtime parameters continues to function correctly.
+  job:
+    replacement_parameters:
+      replaceme: moocow
+  outputs:
+    out1:
+      metadata:
+        name: 'moocow name'
+    out2:
+      metadata:
+        name: 'moocow name 2'
diff --git a/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf.yml b/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf.yml
new file mode 100644
index 000000000000..834721c3cc2b
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_legacy.gxwf.yml
@@ -0,0 +1,17 @@
+class: GalaxyWorkflow
+inputs: {}
+outputs:
+  out1:
+    outputSource: create_2/out_file1
+  out2:
+    outputSource: create_2/out_file2
+steps:
+  create_2:
+    tool_id: create_2
+    state:
+      sleep_time: 0
+    outputs:
+      out_file1:
+        rename: "${replaceme} name"
+      out_file2:
+        rename: "${replaceme} name 2"
diff --git a/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf-tests.yml b/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf-tests.yml
new file mode 100644
index 000000000000..c5152f0e4add
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf-tests.yml
@@ -0,0 +1,13 @@
+- doc: |
+    Test to verify text parameters can be implicitly used as replacement parameters in subworkflows if they are connected up.
+  job:
+    replacemeouter: 
+      value: moocow
+      type: raw
+  outputs:
+    out1:
+      metadata:
+        name: 'moocow name'
+    out2:
+      metadata:
+        name: 'moocow name 2'
diff --git a/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf.yml b/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf.yml
new file mode 100644
index 000000000000..d783195ae584
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_nested.gxwf.yml
@@ -0,0 +1,31 @@
+class: GalaxyWorkflow
+inputs:
+  replacemeouter: text
+outputs:
+  out1:
+    outputSource: nested_workflow/workflow_output_1
+  out2: 
+    outputSource: nested_workflow/workflow_output_2
+steps:
+  nested_workflow:
+    run:
+      class: GalaxyWorkflow
+      inputs:
+        replacemeinner: text
+      outputs:
+        workflow_output_1:
+          outputSource: create_2/out_file1
+        workflow_output_2:
+          outputSource: create_2/out_file2
+      steps:
+        create_2:
+          tool_id: create_2
+          state:
+            sleep_time: 0
+          outputs:
+            out_file1:
+              rename: "${replacemeinner} name"
+            out_file2:
+              rename: "${replacemeinner} name 2"
+    in:
+      replacemeinner: replacemeouter
diff --git a/lib/galaxy_test/workflow/replacement_parameters_text.gxwf-tests.yml b/lib/galaxy_test/workflow/replacement_parameters_text.gxwf-tests.yml
new file mode 100644
index 000000000000..b76353ae9325
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_text.gxwf-tests.yml
@@ -0,0 +1,13 @@
+- doc: |
+    Test to verify text parameters can be implicitly used as replacement parameters in workflow step PJAs.
+  job:
+    replaceme: 
+      value: moocow
+      type: raw
+  outputs:
+    out1:
+      metadata:
+        name: 'moocow name'
+    out2:
+      metadata:
+        name: 'moocow name 2'
diff --git a/lib/galaxy_test/workflow/replacement_parameters_text.gxwf.yml b/lib/galaxy_test/workflow/replacement_parameters_text.gxwf.yml
new file mode 100644
index 000000000000..18f73aadb699
--- /dev/null
+++ b/lib/galaxy_test/workflow/replacement_parameters_text.gxwf.yml
@@ -0,0 +1,18 @@
+class: GalaxyWorkflow
+inputs:
+  replaceme: text
+outputs:
+  out1:
+    outputSource: create_2/out_file1
+  out2:
+    outputSource: create_2/out_file2
+steps:
+  create_2:
+    tool_id: create_2
+    state:
+      sleep_time: 0
+    outputs:
+      out_file1:
+        rename: "${replaceme} name"
+      out_file2:
+        rename: "${replaceme} name 2"
diff --git a/lib/galaxy_test/workflow/tests.py b/lib/galaxy_test/workflow/tests.py
new file mode 100644
index 000000000000..f0702fb8240b
--- /dev/null
+++ b/lib/galaxy_test/workflow/tests.py
@@ -0,0 +1,136 @@
+import glob
+import os
+import tempfile
+from pathlib import Path
+
+import pytest
+import requests
+import yaml
+from gxformat2.yaml import ordered_load
+
+from galaxy.tool_util.parser.interface import TestCollectionOutputDef
+from galaxy.tool_util.verify import verify_file_contents_against_dict
+from galaxy.tool_util.verify.interactor import (
+    compare_expected_metadata_to_api_response,
+    get_metadata_to_test,
+    verify_collection,
+)
+from galaxy_test.api._framework import ApiTestCase
+from galaxy_test.base.populators import (
+    DatasetCollectionPopulator,
+    DatasetPopulator,
+    RunJobsSummary,
+    WorkflowPopulator,
+)
+
+SCRIPT_DIRECTORY = os.path.abspath(os.path.dirname(__file__))
+
+
+def find_workflows():
+    return [Path(p) for p in glob.glob(f"{SCRIPT_DIRECTORY}/*.gxwf.yml")]
+
+
+def pytest_generate_tests(metafunc):
+    parameter_combinations = []
+    test_ids = []
+    for workflow_path in find_workflows():
+        for index, test_job in enumerate(_test_jobs(workflow_path)):
+            parameter_combinations.append([workflow_path, test_job])
+            workflow_test_name = workflow_path.name[0 : -len(".gxwf.yml")]
+            test_ids.append(f"{workflow_test_name}_{index}")
+    if "workflow_path" in metafunc.fixturenames:
+        metafunc.parametrize("workflow_path,test_job", parameter_combinations, ids=test_ids)
+
+
+class TestWorkflow(ApiTestCase):
+    framework_tool_and_types = True
+
+    def setUp(self):
+        super().setUp()
+        self.workflow_populator = WorkflowPopulator(self.galaxy_interactor)
+        self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
+        self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
+
+    @pytest.mark.workflow
+    def test_workflow(self, workflow_path: Path, test_job):
+        with workflow_path.open() as f:
+            yaml_content = ordered_load(f)
+        with self.dataset_populator.test_history() as history_id:
+            run_summary = self.workflow_populator.run_workflow(
+                yaml_content,
+                test_data=test_job["job"],
+                history_id=history_id,
+            )
+            self._verify(run_summary, test_job["outputs"])
+
+    def _verify(self, run_summary: RunJobsSummary, output_definitions):
+        for output_name, output_definition in output_definitions.items():
+            self._verify_output(run_summary, output_name, output_definition)
+
+    def _verify_output(self, run_summary: RunJobsSummary, output_name, test_properties):
+        is_collection_test = "elements" in test_properties
+        item_label = f"Output named {output_name}"
+
+        def get_filename(name):
+            return tempfile.NamedTemporaryFile(prefix=f"gx_workflow_framework_test_file_{output_name}", delete=False)
+
+        def verify_dataset(dataset: dict, test_properties: dict):
+            output_content = self.dataset_populator.get_history_dataset_content(
+                run_summary.history_id, dataset=dataset, type="bytes"
+            )
+            verify_file_contents_against_dict(get_filename, _get_location, item_label, output_content, test_properties)
+            metadata = get_metadata_to_test(test_properties)
+            if metadata:
+                dataset_details = self.dataset_populator.get_history_dataset_details(
+                    run_summary.history_id, content_id=dataset["id"]
+                )
+                compare_expected_metadata_to_api_response(metadata, dataset_details)
+
+        if is_collection_test:
+            test_properties["name"] = output_name
+            # setup preferred name "elements" in accordance with work in https://github.com/galaxyproject/planemo/pull/1417
+            test_properties["element_tests"] = test_properties["elements"]
+            output_def = TestCollectionOutputDef.from_dict(test_properties)
+
+            invocation_details = self.workflow_populator.get_invocation(run_summary.invocation_id, step_details=True)
+            assert output_name in invocation_details["output_collections"]
+            test_output = invocation_details["output_collections"][output_name]
+            output_collection = self.dataset_populator.get_history_collection_details(
+                run_summary.history_id, content_id=test_output["id"]
+            )
+
+            def verify_dataset_element(element, test_properties, element_outfile):
+                hda = element["object"]
+                verify_dataset(hda, test_properties)
+
+            verify_collection(output_def, output_collection, verify_dataset_element)
+        else:
+            test_properties["name"] = output_name
+            invocation_details = self.workflow_populator.get_invocation(run_summary.invocation_id, step_details=True)
+            assert output_name in invocation_details["outputs"]
+            test_output = invocation_details["outputs"][output_name]
+            verify_dataset(test_output, test_properties)
+
+
+def _test_jobs(workflow_path: Path) -> list:
+    test_path = _workflow_test_path(workflow_path)
+    with test_path.open() as f:
+        jobs = yaml.safe_load(f)
+    return jobs
+
+
+def _workflow_test_path(workflow_path: Path) -> Path:
+    base_name = workflow_path.name[0 : -len(".gxwf.yml")]
+    test_path = workflow_path.parent / f"{base_name}.gxwf-tests.yml"
+    return test_path
+
+
+def _get_location(location: str) -> str:
+    data_file = tempfile.NamedTemporaryFile(prefix="gx_workflow_framework_test_file_", delete=False)
+    with requests.get(location, stream=True) as r:
+        r.raise_for_status()
+
+        for chunk in r.iter_content():
+            if chunk:
+                data_file.write(chunk)
+        return data_file.name
diff --git a/lib/galaxy_test/workflow/zip_collection.gxwf-tests.yml b/lib/galaxy_test/workflow/zip_collection.gxwf-tests.yml
new file mode 100644
index 000000000000..e0e4ffc5ba69
--- /dev/null
+++ b/lib/galaxy_test/workflow/zip_collection.gxwf-tests.yml
@@ -0,0 +1,10 @@
+- doc: | 
+    Test simple use of __ZIP_COLLECTION__ in a workflow.
+  job:
+    test_input_1: "samp1\t10.0\nsamp2\t20.0\n"
+    test_input_2: "samp1\t20.0\nsamp2\t40.0\n"
+  outputs:
+    out:
+      asserts:
+      - that: has_text
+        text: "samp1\t10.0\nsamp2\t20.0\nsamp1\t20.0\nsamp2\t40.0"
diff --git a/lib/galaxy_test/workflow/zip_collection.gxwf.yml b/lib/galaxy_test/workflow/zip_collection.gxwf.yml
new file mode 100644
index 000000000000..151e0ffd1aa6
--- /dev/null
+++ b/lib/galaxy_test/workflow/zip_collection.gxwf.yml
@@ -0,0 +1,21 @@
+class: GalaxyWorkflow
+inputs:
+  test_input_1: data
+  test_input_2: data
+outputs:
+  out:
+    outputSource: concat_pair/out1
+steps:
+  first_cat:
+    tool_id: cat
+    in:
+      input1: test_input_1
+  zip_it:
+    tool_id: "__ZIP_COLLECTION__"
+    in:
+      input_forward: first_cat/out_file1
+      input_reverse: test_input_2
+  concat_pair:
+    tool_id: collection_paired_test
+    in:
+      f1: zip_it/output
diff --git a/run_tests.sh b/run_tests.sh
index 2e04eab5eee9..c5b6104a5827 100755
--- a/run_tests.sh
+++ b/run_tests.sh
@@ -14,6 +14,8 @@ cat <<EOF
 '${0##*/} -main'                    for running tests of tools shipped with Galaxy
 '${0##*/} -framework'               for running through example tool tests testing framework features in test/functional/tools"
 '${0##*/} -framework -id toolid'    for testing one framework tool (in test/functional/tools/) with id 'toolid'
+'${0##*/} -framework-workflows'      for running through workflow tests to test correctness of workflow evaluation
+'${0##*/} -framework-workflows -id workflow_name' test single workflow framework test
 '${0##*/} -data_managers -id data_manager_id'    for testing one Data Manager with id 'data_manager_id'
 '${0##*/} -unit'                    for running all unit tests (doctests and tests in test/unit)
 '${0##*/} -unit (test_selector)'    for running unit tests on specified test path (using pytest selector syntax)
@@ -400,13 +402,19 @@ do
           export GALAXY_TEST_USER_API_KEY=$2
           shift 2
           ;;
-      -f|-framework|--framework)
+      -f|-framework|--framework|--framework-tools)
           GALAXY_TEST_TOOL_CONF="test/functional/tools/sample_tool_conf.xml"
           marker="tool"
           report_file="run_framework_tests.html"
           framework_test=1
           shift 1
           ;;
+      -w|--framework-workflows)
+          marker="workflow"
+          report_file="run_framework_workflows_tests.html"
+          framework_workflows_test=1
+          shift 1
+          ;;
       -d|-data_managers|--data_managers)
           marker="data_manager"
           report_file="run_data_managers_tests.html"
@@ -583,6 +591,13 @@ if [ -n "$framework_test" ] || [ -n "$data_managers_test" ] ; then
         selector=""
     fi
     extra_args="test/functional/test_toolbox_pytest.py $selector"
+elif [ -n "$framework_workflows_test" ]; then
+    if [ -n "$test_id" ]; then
+        selector="-k $test_id"
+    else
+        selector=""
+    fi
+    extra_args="lib/galaxy_test/workflow/tests.py $selector"
 elif [ -n "$selenium_script" ]; then
     extra_args="$selenium_script"
 elif [ -n "$toolshed_script" ]; then
diff --git a/test/functional/tools/sample_datatypes_conf.xml b/test/functional/tools/sample_datatypes_conf.xml
index 802c286be68f..f392b35f2eac 100644
--- a/test/functional/tools/sample_datatypes_conf.xml
+++ b/test/functional/tools/sample_datatypes_conf.xml
@@ -48,6 +48,7 @@
     <datatype extension="bgzip" type="galaxy.datatypes.binary:Binary" subclass="true" />
     <datatype extension="vcf_bgzip" type="galaxy.datatypes.tabular:VcfGz" display_in_upload="true"/>
     <datatype extension="html" type="galaxy.datatypes.text:Html" mimetype="text/html"/>
+    <datatype extension="expression.json" type="galaxy.datatypes.text:ExpressionJson" display_in_upload="true"/>
     <datatype extension="data_manager_json" type="galaxy.datatypes.text:DataManagerJson" mimetype="application/json" subclass="true" display_in_upload="false"/>
     <datatype extension="data" type="galaxy.datatypes.data:Data" mimetype="application/octet-stream" max_optional_metadata_filesize="1048576" />
     <datatype extension="binary" type="galaxy.datatypes.binary:Binary" mimetype="application/octet-stream" max_optional_metadata_filesize="1048576" />