diff --git a/core/models/cosmoscope.js b/core/models/cosmoscope.js
index 1c434992..c1bdd322 100644
--- a/core/models/cosmoscope.js
+++ b/core/models/cosmoscope.js
@@ -323,7 +323,7 @@ class Cosmoscope extends Graph {
 
         const fileId = file.metas['id'] || file.metas['title'].toLowerCase();
 
-        quotesWithContexts.forEach(({ id, contexts }) => {
+        quotesWithContexts.forEach(({ id, type, contexts }) => {
           if (!bibliography.library[id]) return;
 
           if (referenceRecords.has(id)) {
@@ -333,19 +333,29 @@ class Cosmoscope extends Graph {
             referenceRecords.set(id, {
               contexts,
               targets: new Set([fileId]),
+              type,
             });
           }
         });
       }
     }
 
-    referenceRecords.forEach(({ targets, contexts }, key) => {
+    referenceRecords.forEach(({ targets, contexts, type }, key) => {
       nodes.push(
         new Node(key, bibliography.library[key]['title'] || '', [opts['references_type_label']]),
       );
       Array.from(targets).forEach((id) =>
         links.push(
-          new Link(undefined, contexts, 'undefined', undefined, undefined, undefined, id, key),
+          new Link(
+            undefined,
+            contexts,
+            type || 'undefined',
+            undefined,
+            undefined,
+            undefined,
+            id,
+            key,
+          ),
         ),
       );
     });
diff --git a/core/utils/citeExtractor.js b/core/utils/citeExtractor.js
index 7ce64377..30e3992a 100644
--- a/core/utils/citeExtractor.js
+++ b/core/utils/citeExtractor.js
@@ -168,6 +168,8 @@ export default function extractCitations(markdown) {
           thisCitation['suppress-author'] = rawPrefix.trim().endsWith('-');
           if (thisCitation['suppress-author']) {
             thisCitation.prefix = rawPrefix.substring(0, rawPrefix.trim().length - 1).trim();
+          } else if (rawPrefix.trim().endsWith(':')) {
+            thisCitation.type = rawPrefix.trim().slice(0, -1);
           } else {
             thisCitation.prefix = rawPrefix.trim();
           }
diff --git a/core/utils/citeExtractor.spec.js b/core/utils/citeExtractor.spec.js
index 011d07f8..29c83f30 100644
--- a/core/utils/citeExtractor.spec.js
+++ b/core/utils/citeExtractor.spec.js
@@ -306,6 +306,22 @@ const tests = [
     ],
     description: 'extracts an in-text citation with a URL as citekey and optional locator/suffix',
   },
+  {
+    input: 'Blah blah [agreesWith: @doe99; about: @smith2000].',
+    expected: [
+      {
+        from: 10,
+        to: 49,
+        composite: false,
+        source: '[agreesWith: @doe99; about: @smith2000]',
+        citations: [
+          { ...defaults, id: 'doe99', type: 'agreesWith' },
+          { ...defaults, id: 'smith2000', type: 'about' },
+        ],
+      },
+    ],
+    description: 'extracts regular citation with type',
+  },
   // The next tests check that locators without explicit page number are detected
   {
     input: '@{https://example.com/bib?name=foobar&date=2000} [33] says blah.',
diff --git a/core/utils/quoteIdsWithContexts.js b/core/utils/quoteIdsWithContexts.js
index 1d6700d6..63720ca2 100644
--- a/core/utils/quoteIdsWithContexts.js
+++ b/core/utils/quoteIdsWithContexts.js
@@ -12,13 +12,14 @@ export default function quoteIdsWithContexts(markdown) {
 
   extractParaphs(markdown).forEach((paraph) => {
     extractCitations(paraph).forEach((result) => {
-      result.citations.forEach(({ id }) => {
+      result.citations.forEach(({ id, type }) => {
         if (quotes[id]) {
           quotes[id].contexts.add(paraph);
         } else {
           quotes[id] = {
             contexts: new Set([paraph]),
             id,
+            type,
           };
         }
       });
diff --git a/core/utils/quoteIdsWithContexts.spec.js b/core/utils/quoteIdsWithContexts.spec.js
index 42ef52b3..5a50ee7a 100644
--- a/core/utils/quoteIdsWithContexts.spec.js
+++ b/core/utils/quoteIdsWithContexts.spec.js
@@ -20,6 +20,20 @@ Second paragraph quoting @doe99.`;
     ]);
   });
 
+  it('should parse quote with type', () => {
+    const text = 'First paragraph quoting [agreesWith: @smith04].';
+
+    const result = quotesWithContexts(text);
+
+    expect(result).toEqual([
+      expect.objectContaining({
+        contexts: ['First paragraph quoting [agreesWith: @smith04].'],
+        id: 'smith04',
+        type: 'agreesWith',
+      }),
+    ]);
+  });
+
   it('should parse many quotes per paragraph', () => {
     const text = `First paragraph quoting @smith04.
 
diff --git a/e2e/citeproc/config.yml b/e2e/citeproc/config.yml
index 19b208b5..4e67fbd0 100644
--- a/e2e/citeproc/config.yml
+++ b/e2e/citeproc/config.yml
@@ -28,6 +28,9 @@ link_types:
   adjacent concept:
     stroke: dotted
     color: "#e1e1e1"
+  agreesWith:
+    stroke: simple
+    color: "green"
 references_as_nodes: true
 references_type_label: "reference"
 record_filters: []
diff --git a/e2e/records/tools-for-thought.md b/e2e/records/tools-for-thought.md
index 6252a2d8..d2bf81a9 100644
--- a/e2e/records/tools-for-thought.md
+++ b/e2e/records/tools-for-thought.md
@@ -6,6 +6,6 @@ tag:
   - quotes
 ---
 
-« Tools to augment human intelligence » [@engelbart1962; quoted by @matuschak2019].
+« Tools to augment human intelligence » [agreesWith: @engelbart1962; quoted by @matuschak2019].
 
 ![alt text](otlet.jpg "Title")
\ No newline at end of file