diff --git a/.github/workflows/conventional-prs.yml b/.github/workflows/conventional-prs.yml
index d4c95052..d45ea2e8 100644
--- a/.github/workflows/conventional-prs.yml
+++ b/.github/workflows/conventional-prs.yml
@@ -16,4 +16,4 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         with:
-          validateSingleCommit: true
\ No newline at end of file
+          validateSingleCommit: true
diff --git a/.github/workflows/sphinx-docs.yml b/.github/workflows/sphinx-docs.yml
index 0f515149..1fdf5d44 100644
--- a/.github/workflows/sphinx-docs.yml
+++ b/.github/workflows/sphinx-docs.yml
@@ -5,7 +5,7 @@ on:
   pull_request:
     paths:
       - 'docs/**'
-  
+
 permissions:
     contents: write
 jobs:
@@ -15,10 +15,10 @@ jobs:
     steps:
       # https://github.com/actions/checkout
       - uses: actions/checkout@v4
-      
+
       # https://github.com/actions/setup-python
       - uses: actions/setup-python@v4
-      
+
       # https://github.com/snok/install-poetry
       - name: Install Poetry
         uses: snok/install-poetry@v1
@@ -37,10 +37,10 @@ jobs:
     steps:
       # https://github.com/actions/checkout
       - uses: actions/checkout@v4
-      
+
       # https://github.com/actions/setup-python
       - uses: actions/setup-python@v4
-      
+
       # https://github.com/snok/install-poetry
       - name: Install Poetry
         uses: snok/install-poetry@v1
diff --git a/README.md b/README.md
index 6090288a..32d8773d 100644
--- a/README.md
+++ b/README.md
@@ -10,7 +10,7 @@ Scientific code repositories contain valuable metadata which can be used to enri
 
 ----------------------------------------------------------------------
 
-Using Gimie: easy peasy, it's a 3 step process. 
+Using Gimie: easy peasy, it's a 3 step process.
 
 ## STEP 1: Installation
 
@@ -37,11 +37,11 @@ docker run -e ACCESS_TOKEN=$ACCESS_TOKEN ghcr.io/sdsc-ord/gimie:latest gimie dat
 
 ## STEP 2 : Set your credentials
 
-In order to access the github api, you need to provide a github token with the `read:org` scope. 
+In order to access the github api, you need to provide a github token with the `read:org` scope.
 
 ### A. Create access tokens
 
-New to access tokens? Or don't know how to get your Github / Gitlab token ? 
+New to access tokens? Or don't know how to get your Github / Gitlab token ?
 
 Have no fear, see
 [here for Github tokens](https://docs.github.com/en/enterprise-server@3.4/authentication/keeping-your-account-and-data-secure/creating-a-personal-access-token) and [here for Gitlab tokens](https://docs.gitlab.com/ee/user/profile/personal_access_tokens.html).
@@ -68,7 +68,7 @@ export GITLAB_TOKEN=
 ```shell
 gimie data https://github.com/numpy/numpy
 ```
-(want a Gitlab project instead? Just replace the URL in the command line) 
+(want a Gitlab project instead? Just replace the URL in the command line)
 
 ### As a python library
 
@@ -132,7 +132,7 @@ run checks:
 ```shell
 make check
 ```
-for an easier use Github/Gitlab APIs, place your access tokens in the `.env` file: (and don't worry, the `.gitignore` will ignore them when you push to GitHub) 
+for an easier use Github/Gitlab APIs, place your access tokens in the `.env` file: (and don't worry, the `.gitignore` will ignore them when you push to GitHub)
 
 ```
 cp .env.dist .env
diff --git a/gimie/cli.py b/gimie/cli.py
index 2a34f6aa..a677e0d9 100644
--- a/gimie/cli.py
+++ b/gimie/cli.py
@@ -72,7 +72,8 @@ def data(
 
     The output is sent to stdout, and turtle is used as the default serialization format."""
     proj = Project(url, base_url=base_url)
-    print(proj.serialize(format=format))
+    repo_meta = proj.extract()
+    print(repo_meta.serialize(format=format))
 
 
 @app.command()
diff --git a/gimie/models.py b/gimie/models.py
index 4a2abb65..1dbaded4 100644
--- a/gimie/models.py
+++ b/gimie/models.py
@@ -17,11 +17,13 @@
 """Data models to represent nodes in the graph generated by gimie."""
 from __future__ import annotations
 from dataclasses import dataclass, field
+from datetime import datetime
 import datetime
-from typing import Optional, List
+from typing import List, Optional, Union
 
 from calamus.schema import JsonLDSchema
 from calamus import fields
+from rdflib import Graph
 
 from gimie.graph.namespaces import SDO
 
@@ -42,7 +44,7 @@ class Release:
     """
 
     tag: str = field(compare=False)
-    date: datetime.datetime = field(compare=True)
+    date: datetime = field(compare=True)
     commit_hash: str = field(compare=False)
 
 
@@ -103,3 +105,76 @@ class PersonSchema(JsonLDSchema):
     class Meta:
         rdf_type = SDO.Person
         model = Person
+
+
+@dataclass
+class Repository:
+    """This class represents a git repository.
+    It does not contain any information about the content of the repository.
+    See https://schema.org/SoftwareSourceCode
+    """
+
+    url: str
+    name: str
+
+    authors: Optional[List[Union[Organization, Person]]] = None
+    contributors: Optional[List[Person]] = None
+    date_created: Optional[datetime] = None
+    date_modified: Optional[datetime] = None
+    date_published: Optional[datetime] = None
+    description: Optional[str] = None
+    download_url: Optional[str] = None
+    identifier: Optional[str] = None
+    keywords: Optional[List[str]] = None
+    licenses: Optional[List[str]] = None
+    parent_repository: Optional[str] = None
+    prog_langs: Optional[List[str]] = None
+    version: Optional[str] = None
+
+    @property
+    def _id(self) -> str:
+        """Unique identifier for the repository."""
+        return self.url
+
+    def to_graph(self) -> Graph:
+        """Convert repository to RDF graph."""
+        jd = RepositorySchema().dumps(self)
+        g: Graph = Graph().parse(format="json-ld", data=str(jd))
+        g.bind("schema", SDO)
+        return g
+
+    def serialize(self, format: str = "ttl", **kwargs) -> str:
+        """Serialize the RDF graph representing the instance."""
+        return self.to_graph().serialize(format=format, **kwargs)  # type: ignore
+
+    def jsonld(self) -> str:
+        """Alias for jsonld serialization."""
+        return self.serialize(format="json-ld")
+
+
+class RepositorySchema(JsonLDSchema):
+    """This defines the schema used for json-ld serialization."""
+
+    _id = fields.Id()
+    authors = fields.Nested(
+        SDO.author, [PersonSchema, OrganizationSchema], many=True
+    )
+    contributors = fields.Nested(SDO.contributor, PersonSchema, many=True)
+    date_created = fields.Date(SDO.dateCreated)
+    date_modified = fields.Date(SDO.dateModified)
+    date_published = fields.Date(SDO.datePublished)
+    description = fields.String(SDO.description)
+    download_url = fields.IRI(SDO.downloadUrl)
+    identifier = fields.String(SDO.identifier)
+    keywords = fields.List(SDO.keywords, fields.String)
+    licenses = fields.List(SDO.license, fields.IRI)
+    name = fields.String(SDO.name)
+    parent_repository = fields.IRI(SDO.isBasedOn)
+    prog_langs = fields.List(SDO.programmingLanguage, fields.String)
+    url = fields.IRI(SDO.codeRepository)
+    version = fields.String(SDO.version)
+
+    class Meta:
+        rdf_type = SDO.SoftwareSourceCode
+        model = Repository
+        add_value_types = False
diff --git a/gimie/project.py b/gimie/project.py
index 59b3033e..f53601b5 100644
--- a/gimie/project.py
+++ b/gimie/project.py
@@ -25,6 +25,7 @@
 from urllib.parse import urlparse
 
 from gimie.graph.operations import combine_graphs
+from gimie.models import Repository
 from gimie.utils import validate_url
 from gimie.sources import SOURCES
 from gimie.sources.abstract import Extractor
@@ -48,6 +49,7 @@ class Project:
     Examples
     --------
     >>> proj = Project("https://github.com/SDSC-ORD/gimie")
+    >>> assert isinstance(proj.extract(), Graph)
     """
 
     def __init__(
@@ -63,24 +65,16 @@ def __init__(
         self._cloned = False
         if validate_url(path):
             self.url = path
-            # We only need to clone a remote project
-            # if a local extractor is enabled
-            if any(map(is_local_source, sources)):
-                self.project_dir = self.clone(path)
         else:
             self.project_dir = path
 
         self.extractors = self.get_extractors(sources)
-        for ex in self.extractors:
-            ex.extract()
 
-    def clone(self, url: str) -> str:
-        """Clone target url in a new temporary directory"""
-        target_dir = TemporaryDirectory().name
-        cloned = git.Repo.clone_from(url, target_dir)  # type: ignore
-        self._cloned = True
-
-        return str(cloned.working_tree_dir)
+    def extract(self) -> Graph:
+        repos = [ex.extract() for ex in self.extractors]
+        graphs = [repo.to_graph() for repo in repos]
+        graph = combine_graphs(*graphs)
+        return graph
 
     def get_extractors(self, sources: Iterable[str]) -> List[Extractor]:
 
@@ -96,29 +90,6 @@ def get_extractors(self, sources: Iterable[str]) -> List[Extractor]:
 
         return extractors
 
-    def to_graph(self) -> Graph:
-        graphs = map(lambda ex: ex.to_graph(), self.extractors)
-        combined_graph = combine_graphs(*graphs)
-        return combined_graph
-
-    def serialize(self, format: str = "ttl", **kwargs):
-        return self.to_graph().serialize(format=format, **kwargs)
-
-    def cleanup(self):
-        """Recursively delete the project. Only works
-        for remote (i.e. cloned) projects."""
-        try:
-            tempdir = gettempdir()
-            if self.project_dir is not None:
-                in_temp = self.project_dir.startswith(tempdir)
-                if self._cloned and in_temp:
-                    shutil.rmtree(self.project_dir)
-        except AttributeError:
-            pass
-
-    def __del__(self):
-        self.cleanup()
-
 
 def split_git_url(url) -> Tuple[str, str]:
     base_url = urlparse(url).scheme + "://" + urlparse(url).netloc
diff --git a/gimie/sources/abstract.py b/gimie/sources/abstract.py
index d46d5a00..8945d335 100644
--- a/gimie/sources/abstract.py
+++ b/gimie/sources/abstract.py
@@ -18,11 +18,11 @@
 from abc import ABC, abstractmethod
 from typing import List, Optional
 
-
-from rdflib import Graph
 from urllib.parse import urlparse
-from gimie.sources.common.license import get_license_url, is_license_path
+
 from gimie.io import Resource
+from gimie.models import Repository
+from gimie.sources.common.license import get_license_url, is_license_path
 
 
 class Extractor(ABC):
@@ -44,32 +44,14 @@ def __init__(
         self.local_path = local_path
 
     @abstractmethod
-    def extract(self):
+    def extract(self) -> Repository:
         """Extract metadata"""
         ...
 
-    @abstractmethod
-    def to_graph(self) -> Graph:
-        """Generate an RDF graph from the instance"""
-        return Graph()
-
     def list_files(self) -> List[Resource]:
         """List all files in the repository HEAD."""
         ...
 
-    def serialize(self, format: str = "ttl", **kwargs) -> str:
-        """Serialize the RDF graph representing the instance."""
-        return self.to_graph().serialize(format=format, **kwargs)  # type: ignore
-
-    def jsonld(self) -> str:
-        """Alias for jsonld serialization."""
-        return self.serialize(format="json-ld")
-
-    @property
-    def _id(self) -> str:
-        """Unique identifier for the repository."""
-        return self.url
-
     @property
     def path(self) -> str:
         """Path to the repository without the base URL."""
@@ -87,6 +69,7 @@ def base(self) -> str:
 
     def _get_licenses(self) -> List[str]:
         """Extracts SPDX License URLs from the repository."""
+        # TODO: Move functionality into a dedicate Parser
         license_files = filter(
             lambda p: is_license_path(p.name), self.list_files()
         )
diff --git a/gimie/sources/git.py b/gimie/sources/git.py
index 09d49561..0b15539c 100644
--- a/gimie/sources/git.py
+++ b/gimie/sources/git.py
@@ -17,18 +17,18 @@
 """Extractor which uses a locally available (usually cloned) repository."""
 from dataclasses import dataclass
 from datetime import datetime
+from functools import cached_property
+import os
+import shutil
+import tempfile
 from typing import List, Optional
 import uuid
 
-from calamus import fields
-from calamus.schema import JsonLDSchema
 import git
 import pydriller
-from rdflib import Graph
 
 from gimie.io import LocalResource
-from gimie.graph.namespaces import SDO
-from gimie.models import Person, PersonSchema
+from gimie.models import Person, Repository
 from gimie.sources.abstract import Extractor
 from pathlib import Path
 
@@ -58,43 +58,59 @@ class GitExtractor(Extractor):
     url: str
     base_url: Optional[str] = None
     local_path: Optional[str] = None
+    _cloned: bool = False
 
-    author: Optional[Person] = None
-    contributors: Optional[List[Person]] = None
-    date_created: Optional[datetime] = None
-    date_modified: Optional[datetime] = None
-    license: Optional[List[str]] = None
-
-    def extract(self):
-        if self.local_path is None:
-            raise ValueError("Local path must be provided for extraction.")
-        self.repository = pydriller.Repository(self.local_path)
+    def extract(self) -> Repository:
         # Assuming author is the first person to commit
-        self.author = self._get_creator()
-        self.contributors = self._get_contributors()
-        self.date_created = self._get_creation_date()
-        self.date_modified = self._get_modification_date()
-        self.license = self._get_licenses()
+        self.repository = self._repo_data
+
+        repo_meta = dict(
+            authors=[self._get_creator()],
+            contributors=self._get_contributors(),
+            date_created=self._get_creation_date(),
+            date_modified=self._get_modification_date(),
+            name=self.path,
+            licenses=self._get_licenses(),
+            url=self.url,
+        )
+
+        return Repository(**repo_meta)  # type: ignore
 
     def list_files(self) -> List[LocalResource]:
-        file_list = []
 
-        if self.local_path is None:
-            return file_list
+        self.repository = self._repo_data
+        file_list = []
 
-        for path in Path(self.local_path).rglob("*"):
+        for path in Path(self.local_path).rglob("*"):  # type: ignore
             if (path.parts[0] == ".git") or not path.is_file():
                 continue
             file_list.append(LocalResource(path))
 
         return file_list
 
-    def to_graph(self) -> Graph:
-        """Generate an RDF graph from the instance"""
-        jd = GitExtractorSchema().dumps(self)
-        g: Graph = Graph().parse(data=str(jd), format="json-ld")
-        g.bind("schema", SDO)
-        return g
+    def __del__(self):
+        """Cleanup the cloned repo if it was cloned and is located in tempdir."""
+        try:
+            # Can't be too careful with temp files
+            tempdir = tempfile.gettempdir()
+            if (
+                self.local_path
+                and self._cloned
+                and self.local_path.startswith(tempdir)
+                and tempdir != os.getcwd()
+            ):
+                shutil.rmtree(self.local_path)
+        except AttributeError:
+            pass
+
+    @cached_property
+    def _repo_data(self) -> pydriller.Repository:
+        """Get the repository data by accessing local data or cloning."""
+        if self.local_path is None:
+            self._cloned = True
+            self.local_path = tempfile.TemporaryDirectory().name
+            git.Repo.clone_from(self.url, self.local_path)  # type: ignore
+        return pydriller.Repository(self.local_path)
 
     def _get_contributors(self) -> List[Person]:
         """Get the authors of the repository."""
@@ -113,12 +129,14 @@ def _get_creation_date(self) -> Optional[datetime]:
 
     def _get_modification_date(self) -> Optional[datetime]:
         """Get the last modification date of the repository."""
+        commit = None
         try:
             for commit in self.repository.traverse_commits():
                 pass
-            return commit.author_date
         except (StopIteration, NameError):
-            return None
+            pass
+        finally:
+            return commit.author_date if commit else None
 
     def _get_creator(self) -> Optional[Person]:
         """Get the creator of the repository."""
@@ -136,24 +154,10 @@ def _dev_to_person(
             uid = str(uuid.uuid4())
         else:
             uid = name.replace(" ", "_").lower()
-        dev_id = f"{self._id}/{uid}"
+        dev_id = f"{self.url}/{uid}"
         return Person(
             _id=dev_id,
             identifier=uid,
             name=name,
             email=email,
         )
-
-
-class GitExtractorSchema(JsonLDSchema):
-    _id = fields.Id()
-    author = fields.Nested(SDO.author, PersonSchema)
-    contributors = fields.Nested(SDO.contributor, PersonSchema, many=True)
-    date_created = fields.Date(SDO.dateCreated)
-    date_modified = fields.Date(SDO.dateModified)
-    license = fields.List(SDO.license, fields.IRI)
-
-    class Meta:
-        rdf_type = SDO.SoftwareSourceCode
-        model = GitExtractor
-        add_value_types = False
diff --git a/gimie/sources/github.py b/gimie/sources/github.py
index c5a763a6..7440cb5f 100644
--- a/gimie/sources/github.py
+++ b/gimie/sources/github.py
@@ -16,9 +16,7 @@
 # limitations under the License.
 from __future__ import annotations
 
-import tempfile
 from dataclasses import dataclass
-from datetime import datetime
 from dateutil.parser import isoparse
 from functools import cached_property
 import os
@@ -26,18 +24,13 @@
 from typing import Any, Dict, List, Optional, Union
 from urllib.parse import urlparse
 from dotenv import load_dotenv
-from calamus import fields
-from calamus.schema import JsonLDSchema
-from rdflib import Graph
 
 from gimie.sources.abstract import Extractor
 from gimie.models import (
     Organization,
-    OrganizationSchema,
     Person,
-    PersonSchema,
+    Repository,
 )
-from gimie.graph.namespaces import SDO
 
 from gimie.io import RemoteResource
 from gimie.sources.common.queries import (
@@ -105,25 +98,6 @@ class GithubExtractor(Extractor):
     local_path: Optional[str] = None
 
     token: Optional[str] = None
-    author: Optional[Union[Organization, Person]] = None
-    contributors: Optional[List[Person]] = None
-    prog_langs: Optional[List[str]] = None
-    download_url: Optional[str] = None
-    description: Optional[str] = None
-    date_created: Optional[datetime] = None
-    date_modified: Optional[datetime] = None
-    date_published: Optional[datetime] = None
-    parent_repository: Optional[str] = None
-    keywords: Optional[List[str]] = None
-    license: Optional[List[str]] = None
-    software_version: Optional[str] = None
-
-    def to_graph(self) -> Graph:
-        """Convert repository to RDF graph."""
-        jd = GithubExtractorSchema().dumps(self)
-        g: Graph = Graph().parse(format="json-ld", data=str(jd))
-        g.bind("schema", SDO)
-        return g
 
     def list_files(self) -> List[RemoteResource]:
         """takes the root repository folder and returns the list of files present"""
@@ -135,37 +109,45 @@ def list_files(self) -> List[RemoteResource]:
         for item in file_dict:
             file = RemoteResource(
                 name=item["name"],
-                url=f'{repo_url}/blob/{defaultbranchref}/{item["path"]}',
+                url=f'{repo_url}/raw/{defaultbranchref}/{item["path"]}',
                 headers=self._set_auth(),
             )
             file_list.append(file)
         return file_list
 
-    def extract(self):
+    def extract(self) -> Repository:
         """Extract metadata from target GitHub repository."""
         data = self._repo_data
-        self.author = self._get_author(data["owner"])
-        self.contributors = self._fetch_contributors()
-        self.description = data["description"]
-        self.date_created = isoparse(data["createdAt"][:-1])
-        self.date_modified = isoparse(data["updatedAt"][:-1])
+
+        repo_meta = dict(
+            authors=[self._get_author(data["owner"])],
+            contributors=self._fetch_contributors(),
+            date_created=isoparse(data["createdAt"][:-1]),
+            date_modified=isoparse(data["updatedAt"][:-1]),
+            description=data["description"],
+            name=self.path,
+            keywords=self._get_keywords(*data["repositoryTopics"]["nodes"]),
+            licenses=self._get_licenses(),
+            url=self.url,
+        )
         if data["parent"]:
-            self.parent_repository = data["parent"]["url"]
+            repo_meta["parent_repository"] = data["parent"]["url"]
+
         if data["latestRelease"]:
-            self.date_published = isoparse(
+            repo_meta["date_published"] = isoparse(
                 data["latestRelease"]["publishedAt"]
             )
 
-        self.license = self._get_licenses()
         if data["primaryLanguage"] is not None:
-            self.prog_langs = [data["primaryLanguage"]["name"]]
-        self.keywords = self._get_keywords(*data["repositoryTopics"]["nodes"])
-        last_release = data["latestRelease"]
-        if last_release is not None:
-            self.version = last_release["name"]
-            self.download_url = (
-                f"{self.url}/archive/refs/tags/{self.version}.tar.gz"
-            )
+            repo_meta["prog_langs"] = [data["primaryLanguage"]["name"]]
+
+        if data["latestRelease"]:
+            version = data["latestRelease"]["name"]
+            download_url = f"{self.url}/archive/refs/tags/{version}.tar.gz"
+            repo_meta["download_url"] = download_url
+            repo_meta["version"] = version
+
+        return Repository(**repo_meta)  # type: ignore
 
     @cached_property
     def _repo_data(self) -> Dict[str, Any]:
@@ -321,29 +303,3 @@ def _get_user(self, node: Dict[str, Any]) -> Person:
             name=node["name"],
             affiliations=orgs,
         )
-
-
-class GithubExtractorSchema(JsonLDSchema):
-    """This defines the schema used for json-ld serialization."""
-
-    _id = fields.Id()
-    path = fields.String(SDO.name)
-    author = fields.Nested(SDO.author, [PersonSchema, OrganizationSchema])
-    contributors = fields.Nested(SDO.contributor, PersonSchema, many=True)
-    prog_langs = fields.List(SDO.programmingLanguage, fields.String)
-    download_url = fields.Raw(SDO.downloadUrl)
-    description = fields.String(SDO.description)
-    date_created = fields.Date(SDO.dateCreated)
-    date_modified = fields.Date(SDO.dateModified)
-    date_published = fields.Date(SDO.datePublished)
-    license = fields.List(SDO.license, fields.IRI)
-    url = fields.IRI(SDO.codeRepository)
-    # NOTE: parent_repository is not available for GitLab's GraphQL API in 2023. Add for GitLab when available
-    parent_repository = fields.IRI(SDO.isBasedOn)
-    keywords = fields.List(SDO.keywords, fields.String)
-    version = fields.String(SDO.version)
-
-    class Meta:
-        rdf_type = SDO.SoftwareSourceCode
-        model = GithubExtractor
-        add_value_types = False
diff --git a/gimie/sources/gitlab.py b/gimie/sources/gitlab.py
index 478a7e34..1d6d212e 100644
--- a/gimie/sources/gitlab.py
+++ b/gimie/sources/gitlab.py
@@ -8,17 +8,11 @@
 from typing import Any, Dict, List, Optional, Union
 from urllib.parse import urlparse
 from dotenv import load_dotenv
-from calamus import fields
-from calamus.schema import JsonLDSchema
-from rdflib import Graph
-import tempfile
-from gimie.graph.namespaces import SDO
 from gimie.io import RemoteResource
 from gimie.models import (
     Organization,
-    OrganizationSchema,
     Person,
-    PersonSchema,
+    Repository,
 )
 from gimie.sources.abstract import Extractor
 from gimie.sources.common.queries import send_graphql_query, send_rest_query
@@ -42,27 +36,6 @@ class GitlabExtractor(Extractor):
     local_path: Optional[str] = None
 
     token: Optional[str] = None
-    name: Optional[str] = None
-    identifier: Optional[str] = None
-    author: Optional[List[Union[Organization, Person]]] = None
-    contributors: Optional[List[Person]] = None
-    prog_langs: Optional[List[str]] = None
-    description: Optional[str] = None
-    date_created: Optional[datetime] = None
-    date_modified: Optional[datetime] = None
-    date_published: Optional[datetime] = None
-    version: Optional[str] = None
-    keywords: Optional[List[str]] = None
-    source_organization: Optional[Organization] = None
-    download_url: Optional[str] = None
-    license: Optional[List[str]] = None
-
-    def to_graph(self) -> Graph:
-        """Convert repository to RDF graph."""
-        jd = GitlabExtractorSchema().dumps(self)
-        g: Graph = Graph().parse(format="json-ld", data=str(jd))
-        g.bind("schema", SDO)
-        return g
 
     def list_files(self) -> List[RemoteResource]:
         """takes the root repository folder and returns the list of files present"""
@@ -78,47 +51,48 @@ def list_files(self) -> List[RemoteResource]:
             file_list.append(file)
         return file_list
 
-    def extract(self):
+    def extract(self) -> Repository:
         """Extract metadata from target Gitlab repository."""
 
         # fetch metadata
         data = self._repo_data
 
-        # Each Gitlab project has a unique identifier (integer)
-        self.identifier = urlparse(data["id"]).path.split("/")[2]
-        # at the moment, Gimie fetches only the group directly related to the project
+        # NOTE(identifier): Each Gitlab project has a unique identifier (integer)
+        # NOTE(author): Fetches only the group directly related to the project
         # the group takes the form: parent/subgroup
-        self.source_organization = self._safe_extract_group(data)
-        self.description = data["description"]
-        self.prog_langs = [lang["name"] for lang in data["languages"]]
-        self.date_created = isoparse(data["createdAt"][:-1])
-        self.date_modified = isoparse(data["lastActivityAt"][:-1])
+
+        # NOTE(contributors): contributors = project members
+        # who are not owners + those that have written merge requests
+        # owners are either multiple individuals or a group. If no user
+        # is marked as owner, contributors are project members or merge
+        # request authors
+        repo_meta = dict(
+            authors=self._safe_extract_author(data),
+            contributors=self._safe_extract_contributors(data),
+            date_created=isoparse(data["createdAt"][:-1]),
+            date_modified=isoparse(data["lastActivityAt"][:-1]),
+            description=data["description"],
+            identifier=urlparse(data["id"]).path.split("/")[2],
+            keywords=data["topics"],
+            licenses=self._get_licenses(),
+            name=self.path,
+            prog_langs=[lang["name"] for lang in data["languages"]],
+            url=self.url,
+        )
+
         if data["releases"]["edges"]:
-            self.date_published = isoparse(
+            repo_meta["date_published"] = isoparse(
                 data["releases"]["edges"][0]["node"]["releasedAt"]
             )
-        self.license = self._get_licenses()
-        self.keywords = data["topics"]
-
-        # Get contributors as the project members that are not owners and those that have written merge requests
-        # owners are either multiple individuals or a group, if not user is marked as owner
-        self.author = self._safe_extract_author(data)
-        # contributors are project members or merge request authors
-        self.contributors = self._safe_extract_contributors(data)
 
         if data["releases"] and (len(data["releases"]["edges"]) > 0):
             # go into releases and take the name from the first node (most recent)
-            self.version = data["releases"]["edges"][0]["node"]["name"]
-            self.download_url = f"{self.url}/-/archive/{self.version}/{self.path.split('/')[-1]}-{self.version}.tar.gz"
-
-    def _safe_extract_group(
-        self, repo: Dict[str, Any]
-    ) -> Optional[Organization]:
-        """Extract the group from a GraphQL repository node if it has one."""
-        if (self.path is not None) and (repo["group"] is not None):
-            repo["group"]["name"] = "/".join(self.path.split("/")[0:-1])
-            return self._get_organization(repo["group"])
-        return None
+            version = data["releases"]["edges"][0]["node"]["name"]
+            repo_meta["version"] = version
+            repo_meta[
+                "download_url"
+            ] = f"{self.url}/-/archive/{version}/{self.path.split('/')[-1]}-{version}.tar.gz"
+        return Repository(**repo_meta)  # type: ignore
 
     def _safe_extract_author(
         self, repo: Dict[str, Any]
@@ -217,11 +191,14 @@ def _repo_data(self) -> Dict[str, Any]:
                 repository {
                     rootRef
                     tree{
-                    blobs{
-                        nodes {
-                            name
-                            webUrl
-                          }}}}
+                        blobs{
+                            nodes {
+                                name
+                                webUrl
+                            }
+                        }
+                    }
+                }
                 releases {
                     edges {
                     node {
@@ -306,31 +283,3 @@ def rest_endpoint(self) -> str:
     @property
     def graphql_endpoint(self) -> str:
         return f"{self.base}/api"
-
-
-class GitlabExtractorSchema(JsonLDSchema):
-    """This defines the schema used for json-ld serialization."""
-
-    _id = fields.Id()
-    path = fields.String(SDO.name)
-    identifier = fields.String(SDO.identifier)
-    source_organization = fields.Nested(SDO.isPartOf, OrganizationSchema)
-    author = fields.Nested(
-        SDO.author, [PersonSchema, OrganizationSchema], many=True
-    )
-    contributors = fields.Nested(SDO.contributor, PersonSchema, many=True)
-    prog_langs = fields.List(SDO.programmingLanguage, fields.String)
-    download_url = fields.Raw(SDO.downloadUrl)
-    description = fields.String(SDO.description)
-    date_created = fields.Date(SDO.dateCreated)
-    date_modified = fields.Date(SDO.dateModified)
-    date_published = fields.Date(SDO.datePublished)
-    license = fields.List(SDO.license, fields.IRI)
-    url = fields.IRI(SDO.codeRepository)
-    keywords = fields.List(SDO.keywords, fields.String)
-    version = fields.String(SDO.version)
-
-    class Meta:
-        rdf_type = SDO.SoftwareSourceCode
-        model = GitlabExtractor
-        add_value_types = False
diff --git a/tests/test_git.py b/tests/test_git.py
index 75548489..8d487cff 100644
--- a/tests/test_git.py
+++ b/tests/test_git.py
@@ -1,11 +1,14 @@
 """Tests for the Gimie command line interface."""
 
 import os
-from gimie.sources.git import GitExtractor
-from gimie.project import Project
 import datetime
+
 import pytest
+
 from gimie.graph.namespaces import GIMIE
+from gimie.io import LocalResource
+from gimie.sources.git import GitExtractor
+from gimie.project import Project
 
 LOCAL_REPOSITORY = os.getcwd()
 RENKU_GITHUB = "https://github.com/SwissDataScienceCenter/renku"
@@ -15,16 +18,16 @@
 @pytest.fixture
 def local_meta():
     """Return metadata for a local repository."""
-    meta = GitExtractor(
+    extractor = GitExtractor(
         "https://github.com/SDSC-ORD/gimie", local_path=LOCAL_REPOSITORY
     )
-    meta.extract()
-    return meta
+    return extractor.extract()
 
 
 def test_git_authors(local_meta):
     """Test part of the authors returned by gimie."""
     contribs = [c.name for c in local_meta.contributors]
+    author = local_meta.authors[0]
     names = [
         "cmdoret",
         "Martin Nathan Tristan Fontanet",
@@ -32,7 +35,7 @@ def test_git_authors(local_meta):
         "sabrinaossey",
     ]
     assert all([n in contribs for n in names])
-    assert local_meta.author.name == "Cyril Matthey-Doret"
+    assert author.name == "Cyril Matthey-Doret"
 
 
 def test_git_creation_date(local_meta):
@@ -47,18 +50,26 @@ def test_git_creation_date(local_meta):
 def test_set_uri():
     meta = GitExtractor(
         "https://example.com/test", local_path=LOCAL_REPOSITORY
-    )
-    meta.extract()
+    ).extract()
     assert meta._id == "https://example.com/test"
 
 
 def test_clone_extract_github():
     """Clone Git repository by setting git extractor
     explicitely and extract metadata locally."""
-    meta = Project(RENKU_GITHUB, sources="git")
+    proj = Project(RENKU_GITHUB, sources="git")
+    assert type(proj.extractors[0]) == GitExtractor
+    proj.extract()
 
 
 def test_clone_unsupported():
     """Instantiate Project from unsupported provider
     with git as default provider"""
-    meta = Project(UNSUPPORTED_PROV)
+    proj = Project(UNSUPPORTED_PROV)
+    assert type(proj.extractors[0]) == GitExtractor
+    proj.extract()
+
+
+def test_git_list_files():
+    files = GitExtractor(UNSUPPORTED_PROV).list_files()
+    assert all(isinstance(f, LocalResource) for f in files)
diff --git a/tests/test_github.py b/tests/test_github.py
index f67e533d..053b8c65 100644
--- a/tests/test_github.py
+++ b/tests/test_github.py
@@ -1,4 +1,5 @@
 # Tests fetching metadata from GitHub repositories with different setups.
+from gimie.io import RemoteResource
 from gimie.sources.github import GithubExtractor
 import pytest
 
@@ -12,7 +13,12 @@
 
 
 @pytest.mark.parametrize("repo", TEST_REPOS)
-def test_github_extractor(repo):
-    meta = GithubExtractor(repo)
-    meta.extract()
+def test_github_extract(repo):
+    meta = GithubExtractor(repo).extract()
     meta.serialize(format="ttl")
+
+
+@pytest.mark.parametrize("repo", TEST_REPOS)
+def test_github_list_files(repo):
+    files = GithubExtractor(repo).list_files()
+    assert all(isinstance(f, RemoteResource) for f in files)
diff --git a/tests/test_gitlab.py b/tests/test_gitlab.py
index 84807a76..8c2b32b8 100644
--- a/tests/test_gitlab.py
+++ b/tests/test_gitlab.py
@@ -1,3 +1,4 @@
+from gimie.io import RemoteResource
 from gimie.sources.gitlab import GitlabExtractor
 import pytest
 
@@ -12,7 +13,13 @@
 
 
 @pytest.mark.parametrize("repo", TEST_REPOS)
-def test_gitlab_extractor(repo):
-    meta = GitlabExtractor(repo)
-    meta.extract()
+def test_gitlab_extract(repo):
+    extractor = GitlabExtractor(repo)
+    meta = extractor.extract()
     meta.serialize(format="ttl")
+
+
+@pytest.mark.parametrize("repo", TEST_REPOS)
+def test_gitlab_list_files(repo):
+    files = GitlabExtractor(repo).list_files()
+    assert all(isinstance(f, RemoteResource) for f in files)
diff --git a/tests/test_output.py b/tests/test_output.py
index 144949b4..85758bba 100644
--- a/tests/test_output.py
+++ b/tests/test_output.py
@@ -22,9 +22,11 @@
 from gimie.project import Project
 
 
-OUT_TTL = Project(
-    "https://github.com/SDSC-ORD/gimie", sources=["github"]
-).serialize(format="ttl")
+OUT_TTL = (
+    Project("https://github.com/SDSC-ORD/gimie", sources=["github"])
+    .extract()
+    .serialize(format="ttl")
+)
 
 
 def test_validate_output_is_linked_data():