diff --git a/docs/plans/006-matcher-readability-cleanup.md b/docs/plans/006-matcher-readability-cleanup.md
new file mode 100644
index 0000000..2bd774b
--- /dev/null
+++ b/docs/plans/006-matcher-readability-cleanup.md
@@ -0,0 +1,54 @@
+# Matcher Readability Cleanup
+
+## Approach
+
+Refactor matcher regular expression construction so the current field, value,
+and removal fragments are named and easier to review. This preserves matcher
+exports, function signatures, capture groups, flags, and matching behavior while
+making later parser-first exploration less risky.
+
+## Pre-implementation
+
+Create branch `refactor/matcher-readability-cleanup` from `main` after merging
+the hardening regression coverage branch.
+
+## Steps
+
+1. `docs/plans/006-matcher-readability-cleanup.md` - add this plan for the
+   behavior-preserving matcher cleanup.
+2. `src/matchers.ts` - name local regex fragments for form-encoded, JSON-like,
+   and escaped JSON-like matchers without changing exported APIs or regex
+   behavior.
+
+## Relevant Files
+
+- `docs/plans/006-matcher-readability-cleanup.md` - new plan for this refactor
+  slice.
+- `src/matchers.ts` - updated matcher construction with named local fragments.
+
+## Verification
+
+1. Run `yarn test`.
+2. Run `yarn test:coverage`.
+3. Run `yarn lint`.
+4. Run `yarn format:check`.
+5. Run `yarn build`.
+6. Re-run workspace diagnostics for touched plan and source files.
+
+## Decisions
+
+**Behavior-preserving only** - this branch keeps current regex semantics, even
+where future work may choose a parser-first approach, so existing consumers see
+no compatibility change.
+
+**Preserve capture groups** - masking relies on `$1` and `$2` replacement groups
+in `stringReplacer`, so local fragment extraction must keep group ordering
+intact.
+
+**Keep regex assembly local** - each matcher keeps its format-specific regex
+pieces near the final `RegExp` construction so reviewers can inspect behavior
+without jumping through a helper layer.
+
+**Keep parser-first work separate** - JSON parsing can affect whitespace,
+serialization, and malformed string handling, so it remains a later exploration
+rather than part of this readability refactor.
diff --git a/src/matchers.ts b/src/matchers.ts
index d2d5b30..df6d1c4 100644
--- a/src/matchers.ts
+++ b/src/matchers.ts
@@ -1,5 +1,7 @@
 import { DataSanitizationMatcher } from './types';
 
+const MATCHER_FLAGS = 'gi';
+
 /**
  * Escapes regular expression metacharacters in a pattern string.
  *
@@ -52,13 +54,20 @@ const formEncodedMatcher: DataSanitizationMatcher = (
   remove = false,
 ) => {
   const escaped = escapePattern(pattern);
+  const fieldName = `\\w*${escaped}\\w*`;
+  const fieldPrefix = `${fieldName}[=:]`;
+  const fieldValue = '[^&]*';
+
   if (remove) {
-    return new RegExp(
-      `&\\w*${escaped}\\w*[=:][^&]*|\\w*${escaped}\\w*[=:][^&]*&?`,
-      'gi',
-    );
+    const removeLeadingField = `&${fieldPrefix}${fieldValue}`;
+    const removeField = `${fieldPrefix}${fieldValue}&?`;
+
+    return new RegExp(`${removeLeadingField}|${removeField}`, MATCHER_FLAGS);
   }
-  return new RegExp(`(\\w*${escaped}\\w*[=:])[^&]*(&|$)`, 'gi');
+
+  const maskField = `(${fieldPrefix})${fieldValue}(&|$)`;
+
+  return new RegExp(maskField, MATCHER_FLAGS);
 };
 
 /**
@@ -95,13 +104,21 @@ const formEncodedMatcher: DataSanitizationMatcher = (
  */
 const jsonMatcher: DataSanitizationMatcher = (pattern, remove = false) => {
   const escaped = escapePattern(pattern);
+  const fieldName = `\\w*${escaped}\\w*`;
+
   if (remove) {
-    return new RegExp(
-      `,\\s*"\\w*${escaped}\\w*"\\s*:\\s*"[^"]*"|"\\w*${escaped}\\w*"\\s*:\\s*"[^"]*"\\s*,?`,
-      'gi',
-    );
+    const fieldPrefix = `"${fieldName}"\\s*:\\s*"`;
+    const fieldValue = '[^"]*"';
+    const removeLeadingField = `,\\s*${fieldPrefix}${fieldValue}`;
+    const removeField = `${fieldPrefix}${fieldValue}\\s*,?`;
+
+    return new RegExp(`${removeLeadingField}|${removeField}`, MATCHER_FLAGS);
   }
-  return new RegExp(`("\\w*${escaped}\\w*"?:\\s*").+?(")`, 'gi');
+
+  const fieldPrefix = `"${fieldName}"?:\\s*"`;
+  const maskField = `(${fieldPrefix}).+?(")`;
+
+  return new RegExp(maskField, MATCHER_FLAGS);
 };
 
 /**
@@ -126,16 +143,20 @@ const escapedJsonMatcher: DataSanitizationMatcher = (
   remove = false,
 ) => {
   const escaped = escapePattern(pattern);
+  const fieldName = `\\w*${escaped}\\w*`;
+  const fieldPrefix = `\\\\"${fieldName}\\\\"\\s*:\\s*\\\\"`;
+
   if (remove) {
-    return new RegExp(
-      `,\\s*\\\\"\\w*${escaped}\\w*\\\\"\\s*:\\s*\\\\"[^\\\\"]*\\\\"|\\\\"\\w*${escaped}\\w*\\\\"\\s*:\\s*\\\\"[^\\\\"]*\\\\"\\s*,?`,
-      'gi',
-    );
+    const fieldValue = '[^\\\\"]*\\\\"';
+    const removeLeadingField = `,\\s*${fieldPrefix}${fieldValue}`;
+    const removeField = `${fieldPrefix}${fieldValue}\\s*,?`;
+
+    return new RegExp(`${removeLeadingField}|${removeField}`, MATCHER_FLAGS);
   }
-  return new RegExp(
-    `(\\\\"\\w*${escaped}\\w*\\\\"\\s*:\\s*\\\\").+?(\\\\")`,
-    'gi',
-  );
+
+  const maskField = `(${fieldPrefix}).+?(\\\\")`;
+
+  return new RegExp(maskField, MATCHER_FLAGS);
 };
 
 const defaultMatchers = [formEncodedMatcher, jsonMatcher, escapedJsonMatcher];