diff --git a/.github/workflows/cli.yml b/.github/workflows/cli.yml
new file mode 100644
index 0000000..cac8c7a
--- /dev/null
+++ b/.github/workflows/cli.yml
@@ -0,0 +1,30 @@
+create-release:
+    runs-on: ubuntu-latest
+    # Use a container with GLIBC 2.17
+    container: quay.io/pypa/manylinux2014_x86_64
+    steps:
+      - name: Show GLIBC
+        run: ldd --version
+      - name: Checkout sources
+        uses: actions/checkout@v4
+
+      - name: Install stable toolchain
+        uses: actions-rs/toolchain@v1
+        with:
+          profile: minimal
+          toolchain: stable
+          override: true
+
+      - uses: Swatinem/rust-cache@v1
+
+      - name: Build
+        uses: actions-rs/cargo@v1
+        with:
+          command: build
+          args: --release --manifest-path cli/Cargo.toml
+      - name: Upload binary
+        uses: actions/upload-artifact@4
+        with: 
+          name: cli-tool
+          path: cli
+
diff --git a/Cargo.toml b/Cargo.toml
index 6718eeb..0b2e1a6 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -30,3 +30,4 @@ reqwest = { version = "0.11", features =  ["blocking", "json"], default-features
 
 [dev-dependencies]
 tempfile = "3.9"
+
diff --git a/python/README.md b/python/README.md
index af8e8b3..052c580 100644
--- a/python/README.md
+++ b/python/README.md
@@ -1,6 +1,6 @@
 # Overview 
 
-pycrate is a python library that is built upon a rust backend for interfacing with [RO-Crates](https://www.researchobject.org/ro-crate/1.1/). It's designed to be maximally flexible with minimal onboarding, allowing you to incorprate it into scrpits/ data pipelines as easily as possible. This also relies on you to have an understanding of the structure of an RO-Crate, but focuses more on the fact that some metadata is better than no metadata. 
+rocraters is a python library that is built upon a rust backend for interfacing with [RO-Crates](https://www.researchobject.org/ro-crate/1.1/). It's designed to be maximally flexible with minimal onboarding, allowing you to incorprate it into scrpits/ data pipelines as easily as possible. This also relies on you to have an understanding of the structure of an RO-Crate, but focuses more on the fact that some metadata is better than no metadata. 
 
 # Build 
 
diff --git a/src/ro_crate/metadata_descriptor.rs b/src/ro_crate/metadata_descriptor.rs
index aea503a..b08870e 100644
--- a/src/ro_crate/metadata_descriptor.rs
+++ b/src/ro_crate/metadata_descriptor.rs
@@ -13,7 +13,6 @@ use serde::{
 };
 use std::collections::HashMap;
 use std::fmt;
-use std::marker::PhantomData;
 
 /// Represents the metadata descriptor for entire RoCrate.
 ///
@@ -140,19 +139,12 @@ pub trait CustomSerialize: Serialize {
 /// `@id`, `@type`, `conformsTo`, and `about` keys to populate the corresponding fields
 /// of `MetadataDescriptor`. All other keys are treated as dynamic properties and are
 /// stored in the `dynamic_entity` hashmap.
-///
-/// # Implementation Details
-/// `MetadataDescriptorVisitor` uses `PhantomData` to correctly implement the `Visitor` trait
-/// without needing an actual instance of `MetadataDescriptor`. This allows the visitor to be
-/// generic over the `MetadataDescriptor` type, ensuring type safety and correctness.
 impl<'de> Deserialize<'de> for MetadataDescriptor {
     fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
     where
         D: Deserializer<'de>,
     {
-        struct MetadataDescriptorVisitor {
-            marker: PhantomData<fn() -> MetadataDescriptor>,
-        }
+        struct MetadataDescriptorVisitor;
 
         impl<'de> Visitor<'de> for MetadataDescriptorVisitor {
             type Value = MetadataDescriptor;
@@ -203,9 +195,7 @@ impl<'de> Deserialize<'de> for MetadataDescriptor {
             }
         }
 
-        deserializer.deserialize_map(MetadataDescriptorVisitor {
-            marker: PhantomData,
-        })
+        deserializer.deserialize_map(MetadataDescriptorVisitor)
     }
 }