diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index d8b1157..2022a2d 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -19,7 +19,7 @@ jobs:
       RUST_BACKTRACE: full
 
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v6
     - uses: dsherret/rust-toolchain-file@v1
     - uses: Swatinem/rust-cache@v2
       with:
@@ -35,32 +35,19 @@ jobs:
       if: matrix.config.kind == 'test_release'
       run: cargo test --release --all-features
 
-      # CARGO PUBLISH
-    - name: Cargo login
-      if: matrix.config.kind == 'test_release' && startsWith(github.ref, 'refs/tags/')
-      run: cargo login ${{ secrets.CRATES_TOKEN }}
-
-    - name: Cargo publish
-      if: matrix.config.kind == 'test_release' && startsWith(github.ref, 'refs/tags/')
-      run: cargo publish
-
   benchmark:
     name: Benchmarks
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v6
     - name: Install latest nightly
       uses: actions-rs/toolchain@v1
       with:
         toolchain: nightly
         override: true
     - name: Cache cargo
-      uses: actions/cache@v4
+      uses: Swatinem/rust-cache@v2
       with:
-        path: |
-          ~/.cargo/registry
-          ~/.cargo/git
-          target
         key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}
     # Run benchmark and stores the output to a file
     - name: Run benchmark
diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml
new file mode 100644
index 0000000..9cf7cb5
--- /dev/null
+++ b/.github/workflows/publish.yml
@@ -0,0 +1,21 @@
+name: publish
+
+on:
+  push:
+    tags:
+      - "*"
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      id-token: write
+    steps:
+      - name: Clone repository
+        uses: actions/checkout@v6
+      - uses: rust-lang/crates-io-auth-action@v1
+        id: auth
+      - run: cargo publish
+        env:
+          CARGO_REGISTRY_TOKEN: ${{ steps.auth.outputs.token }}
diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index 7839356..b78affe 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -20,7 +20,7 @@ jobs:
 
     steps:
       - name: Clone repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           token: ${{ secrets.GH_DPRINTBOT_PAT }}
 
diff --git a/rust-toolchain.toml b/rust-toolchain.toml
index 1be126d..711b3a4 100644
--- a/rust-toolchain.toml
+++ b/rust-toolchain.toml
@@ -1,3 +1,3 @@
 [toolchain]
-channel = "1.89.0"
+channel = "1.92.0"
 components = ["clippy", "rustfmt"]
diff --git a/src/errors.rs b/src/errors.rs
index fcfeff4..abd0ad6 100644
--- a/src/errors.rs
+++ b/src/errors.rs
@@ -14,6 +14,7 @@ pub enum ParseErrorKind {
   ExpectedPlusMinusOrDigitInNumberLiteral,
   ExpectedStringObjectProperty,
   HexadecimalNumbersNotAllowed,
+  ExpectedComma,
   MultipleRootJsonValues,
   SingleQuotedStringsNotAllowed,
   String(ParseStringErrorKind),
@@ -59,6 +60,9 @@ impl std::fmt::Display for ParseErrorKind {
       HexadecimalNumbersNotAllowed => {
         write!(f, "Hexadecimal numbers are not allowed")
       }
+      ExpectedComma => {
+        write!(f, "Expected comma")
+      }
       MultipleRootJsonValues => {
         write!(f, "Text cannot contain more than one JSON value")
       }
diff --git a/src/lib.rs b/src/lib.rs
index 5755a68..70fd3a3 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -94,7 +94,8 @@
 //!
 //! ## Parse Strictly as JSON
 //!
-//! Provide `ParseOptions` and set all the options to false:
+//! By default this library is extremely loose in what it allows parsing. To be strict,
+//! provide `ParseOptions` and set all the options to false:
 //!
 //! ```
 //! use jsonc_parser::parse_to_value;
@@ -106,6 +107,7 @@
 //!   allow_comments: false,
 //!   allow_loose_object_property_names: false,
 //!   allow_trailing_commas: false,
+//!   allow_missing_commas: false,
 //!   allow_single_quoted_strings: false,
 //!   allow_hexadecimal_numbers: false,
 //!   allow_unary_plus_numbers: false,
diff --git a/src/parse_to_ast.rs b/src/parse_to_ast.rs
index a98be25..8b540cd 100644
--- a/src/parse_to_ast.rs
+++ b/src/parse_to_ast.rs
@@ -55,6 +55,8 @@ pub struct ParseOptions {
   pub allow_loose_object_property_names: bool,
   /// Allow trailing commas on object literal and array literal values (defaults to `true`).
   pub allow_trailing_commas: bool,
+  /// Allow missing commas between object properties or array elements (defaults to `true`).
+  pub allow_missing_commas: bool,
   /// Allow single-quoted strings (defaults to `true`).
   pub allow_single_quoted_strings: bool,
   /// Allow hexadecimal numbers like 0xFF (defaults to `true`).
@@ -69,6 +71,7 @@ impl Default for ParseOptions {
       allow_comments: true,
       allow_loose_object_property_names: true,
       allow_trailing_commas: true,
+      allow_missing_commas: true,
       allow_single_quoted_strings: true,
       allow_hexadecimal_numbers: true,
       allow_unary_plus_numbers: true,
@@ -102,6 +105,7 @@ struct Context<'a> {
   collect_comments_as_tokens: bool,
   allow_comments: bool,
   allow_trailing_commas: bool,
+  allow_missing_commas: bool,
   allow_loose_object_property_names: bool,
 }
 
@@ -261,6 +265,7 @@ pub fn parse_to_ast<'a>(
     collect_comments_as_tokens: collect_options.comments == CommentCollectionStrategy::AsTokens,
     allow_comments: parse_options.allow_comments,
     allow_trailing_commas: parse_options.allow_trailing_commas,
+    allow_missing_commas: parse_options.allow_missing_commas,
     allow_loose_object_property_names: parse_options.allow_loose_object_property_names,
   };
   context.scan()?;
@@ -321,13 +326,24 @@ fn parse_object<'a>(context: &mut Context<'a>) -> Result<Object<'a>, ParseError>
     }
 
     // skip the comma
-    if let Some(Token::Comma) = context.scan()? {
-      let comma_range = context.create_range_from_last_token();
-      if let Some(Token::CloseBrace) = context.scan()?
-        && !context.allow_trailing_commas
-      {
-        return Err(context.create_error_for_range(comma_range, ParseErrorKind::TrailingCommasNotAllowed));
+    let after_value_end = context.last_token_end;
+    match context.scan()? {
+      Some(Token::Comma) => {
+        let comma_range = context.create_range_from_last_token();
+        if let Some(Token::CloseBrace) = context.scan()?
+          && !context.allow_trailing_commas
+        {
+          return Err(context.create_error_for_range(comma_range, ParseErrorKind::TrailingCommasNotAllowed));
+        }
+      }
+      Some(Token::String(_) | Token::Word(_) | Token::Number(_)) if !context.allow_missing_commas => {
+        let range = Range {
+          start: after_value_end,
+          end: after_value_end,
+        };
+        return Err(context.create_error_for_range(range, ParseErrorKind::ExpectedComma));
       }
+      _ => {}
     }
   }
 
@@ -567,6 +583,7 @@ mod tests {
         allow_comments: false,
         allow_loose_object_property_names: false,
         allow_trailing_commas: false,
+        allow_missing_commas: false,
         allow_single_quoted_strings: false,
         allow_hexadecimal_numbers: false,
         allow_unary_plus_numbers: false,
@@ -673,4 +690,49 @@ mod tests {
     let number_value = obj.properties[0].value.as_number_lit().unwrap();
     assert_eq!(number_value.value, "+42");
   }
+
+  #[test]
+  fn missing_comma_between_properties() {
+    let text = r#"{
+  "name": "alice"
+  "age": 25
+}"#;
+    let result = parse_to_ast(text, &Default::default(), &Default::default()).unwrap();
+    assert_eq!(
+      result
+        .value
+        .unwrap()
+        .as_object()
+        .unwrap()
+        .get_number("age")
+        .unwrap()
+        .value,
+      "25"
+    );
+
+    // but is strict when strict
+    assert_has_strict_error(text, "Expected comma on line 2 column 18");
+  }
+
+  #[test]
+  fn missing_comma_with_comment_between_properties() {
+    // when comments are allowed but missing commas are not,
+    // should still detect the missing comma after the comment is skipped
+    let result = parse_to_ast(
+      r#"{
+  "name": "alice" // comment here
+  "age": 25
+}"#,
+      &Default::default(),
+      &ParseOptions {
+        allow_comments: true,
+        allow_missing_commas: false,
+        ..Default::default()
+      },
+    );
+    match result {
+      Ok(_) => panic!("Expected error, but did not find one."),
+      Err(err) => assert_eq!(err.to_string(), "Expected comma on line 2 column 18"),
+    }
+  }
 }
diff --git a/src/serde.rs b/src/serde.rs
index 7c21c73..80c5c80 100644
--- a/src/serde.rs
+++ b/src/serde.rs
@@ -3,7 +3,7 @@ use super::ParseOptions;
 use super::errors::ParseError;
 use super::parse_to_ast;
 
-/// Parses a string containing JSONC to a `serde_json::Value.
+/// Parses a string containing JSONC to a `serde_json::Value`.
 ///
 /// Requires the "serde" cargo feature:
 ///
@@ -16,7 +16,10 @@ use super::parse_to_ast;
 /// ```rs
 /// use jsonc_parser::parse_to_serde_value;
 ///
-/// let json_value = parse_to_serde_value(r#"{ "test": 5 } // test"#, &Default::default()).unwrap();
+/// let json_value = parse_to_serde_value(
+///   r#"{ "test": 5 } // test"#,
+///   &Default::default(),
+/// ).unwrap();
 /// ```
 pub fn parse_to_serde_value(text: &str, parse_options: &ParseOptions) -> Result<Option<serde_json::Value>, ParseError> {
   let value = parse_to_ast(