diff --git a/YKanBan.Tests/Export/MarkdownExporterTests.cs b/YKanBan.Tests/Export/MarkdownExporterTests.cs
new file mode 100644
index 0000000..c7f9a75
--- /dev/null
+++ b/YKanBan.Tests/Export/MarkdownExporterTests.cs
@@ -0,0 +1,206 @@
+using System.Globalization;
+using System.Text;
+using YKanBan.Export;
+using YKanBan.Models;
+
+namespace YKanBan.Tests.Export;
+
+/// <summary>
+/// Tests for the Markdown exporter. Mutates the global
+/// <see cref="Resources.Culture"/>, therefore marked do-not-parallelize.
+/// Expected documents are assembled from the same ResX template keys as the
+/// exporter (plus fixed fixture values), so no localized text is hard-coded
+/// here.
+/// </summary>
+[TestClass]
+[DoNotParallelize]
+public class MarkdownExporterTests
+{
+    private static readonly DateTimeOffset ExportedAt = new(2026, 9, 30, 14, 25, 37, TimeSpan.Zero);
+
+    /// <summary>
+    /// Builds a local wall-clock moment, matching the exporter's conversion path.
+    /// </summary>
+    /// <param name="year">The year.</param>
+    /// <param name="month">The month.</param>
+    /// <param name="day">The day.</param>
+    /// <param name="hour">The hour.</param>
+    /// <param name="minute">The minute.</param>
+    /// <returns>The local time.</returns>
+    private static DateTimeOffset Local(int year, int month, int day, int hour, int minute) =>
+        new DateTimeOffset(new DateTime(year, month, day, hour, minute, 0)).ToLocalTime();
+
+    /// <summary>
+    /// Converts a local wall-clock time to Unix seconds.
+    /// </summary>
+    /// <param name="year">The year.</param>
+    /// <param name="month">The month.</param>
+    /// <param name="day">The day.</param>
+    /// <param name="hour">The hour.</param>
+    /// <param name="minute">The minute.</param>
+    /// <returns>The Unix timestamp.</returns>
+    private static long UnixTimeOf(int year, int month, int day, int hour, int minute) =>
+        Local(year, month, day, hour, minute).ToUnixTimeSeconds();
+
+    /// <summary>
+    /// Builds the board fixture: one working column with two cards, one empty column and two tags.
+    /// </summary>
+    /// <returns>The columns, cards and tag/usage tuples.</returns>
+    private static (IReadOnlyList<ColumnModel> Columns, IReadOnlyList<CardModel> Cards, IReadOnlyList<(TagModel, long)> Tags) CreateBoard()
+    {
+        long created = UnixTimeOf(2026, 9, 1, 10, 0);
+        long updated = UnixTimeOf(2026, 9, 28, 18, 30);
+
+        var columns = new List<ColumnModel>
+        {
+            new(1, "In Progress", "The work column", 100, 100),
+            new(2, "Empty", string.Empty, 100, 100),
+        };
+        var bug = new TagModel(1, "bug", RgbColor.Parse("#ff0000"), string.Empty);
+        var ui = new TagModel(2, "ui", RgbColor.Parse("#00ff00"), "Interface work");
+        var cards = new List<CardModel>
+        {
+            new(61, 1, "Fix login", "Body text", created, updated, [bug, ui]),
+            new(62, 1, string.Empty, string.Empty, created, updated, []),
+        };
+        var tags = new List<(TagModel, long)> { (bug, 1), (ui, 1) };
+        return (columns, cards, tags);
+    }
+
+    /// <summary>
+    /// Renders the fixture board with the exporter under the current culture.
+    /// </summary>
+    /// <returns>The produced Markdown document.</returns>
+    private static string ExportBoard()
+    {
+        (IReadOnlyList<ColumnModel> columns, IReadOnlyList<CardModel> cards, IReadOnlyList<(TagModel, long)> tags) = CreateBoard();
+        return MarkdownExporter.Export("MyRepo", columns, cards, tags, ExportedAt);
+    }
+
+    /// <summary>
+    /// Assembles the expected document from the ResX template keys under the current culture.
+    /// </summary>
+    /// <param name="culture">The UI culture.</param>
+    /// <returns>The expected Markdown document.</returns>
+    private static string BuildExpected(CultureInfo culture)
+    {
+        string created = Local(2026, 9, 1, 10, 0).ToString(Resources.Export_CardTimePattern, culture);
+        string updated = Local(2026, 9, 28, 18, 30).ToString(Resources.Export_CardTimePattern, culture);
+
+        var builder = new StringBuilder();
+
+        // Header: workspace name and the export timestamp.
+        builder.AppendLine(string.Format(culture, Resources.Export_WorkspaceHeading, "MyRepo"));
+        builder.AppendLine(string.Format(
+            culture, Resources.Export_ExportedAtLine, ExportedAt.ToString(Resources.Export_TimePattern, culture)));
+        builder.AppendLine();
+
+        // First column with its two cards.
+        builder.AppendLine(string.Format(culture, Resources.Export_ColumnHeading, "In Progress", 2));
+        builder.AppendLine("The work column");
+        builder.AppendLine();
+        builder.AppendLine("### #61 Fix login");
+        builder.AppendLine("Body text");
+        builder.AppendLine(string.Format(culture, Resources.Export_TagsLine, "`bug` `ui`"));
+        builder.AppendLine(string.Format(culture, Resources.Export_CreatedModifiedLine, created, updated));
+        builder.AppendLine();
+        builder.AppendLine("### #62");
+        builder.AppendLine(string.Format(culture, Resources.Export_CreatedModifiedLine, created, updated));
+        builder.AppendLine();
+
+        // Empty column is kept.
+        builder.AppendLine(string.Format(culture, Resources.Export_ColumnHeading, "Empty", 0));
+        builder.AppendLine();
+
+        // Tag summary table.
+        builder.AppendLine(Resources.Export_TagsSummaryHeading);
+        builder.AppendLine(Resources.Export_TagsTableHeader);
+        builder.AppendLine("|---|---|---|---|");
+        builder.AppendLine("| bug | #ff0000 |  | 1 |");
+        builder.AppendLine("| ui | #00ff00 | Interface work | 1 |");
+
+        return builder.ToString();
+    }
+
+    /// <summary>
+    /// Normalizes CRLF line endings so expected and actual compare consistently.
+    /// </summary>
+    /// <param name="text">The text to normalize.</param>
+    /// <returns>The text with normalized line endings.</returns>
+    private static string Normalize(string text) => text.Replace("\r\n", "\n");
+
+    /// <summary>
+    /// Asserts that the exporter output matches the resource-built expected document.
+    /// </summary>
+    /// <param name="culture">The UI culture to test.</param>
+    private static void AssertOutputMatches(CultureInfo culture)
+    {
+        CultureInfo? original = Resources.Culture;
+        try
+        {
+            Resources.Culture = culture;
+            Assert.AreEqual(Normalize(BuildExpected(culture)), Normalize(ExportBoard()));
+        }
+        finally
+        {
+            Resources.Culture = original;
+        }
+    }
+
+    [TestMethod]
+    public void EnglishOutputMatchesTemplate()
+    {
+        AssertOutputMatches(new CultureInfo("en"));
+    }
+
+    [TestMethod]
+    public void ChineseOutputMatchesTemplate()
+    {
+        AssertOutputMatches(new CultureInfo("zh-Hans"));
+    }
+
+    [TestMethod]
+    public void FileNameFollowsLanguage()
+    {
+        CultureInfo? original = Resources.Culture;
+        try
+        {
+            Resources.Culture = new CultureInfo("en");
+            Assert.AreEqual(
+                string.Format(Resources.Culture, Resources.Export_FileNamePattern, "MyRepo", ExportedAt),
+                MarkdownExporter.BuildFileName("MyRepo", ExportedAt));
+
+            Resources.Culture = new CultureInfo("zh-Hans");
+            Assert.AreEqual(
+                string.Format(Resources.Culture, Resources.Export_FileNamePattern, "MyRepo", ExportedAt),
+                MarkdownExporter.BuildFileName("MyRepo", ExportedAt));
+        }
+        finally
+        {
+            Resources.Culture = original;
+        }
+    }
+
+    [TestMethod]
+    public void TagTableCellsEscapePipesAndLineBreaks()
+    {
+        CultureInfo? original = Resources.Culture;
+        try
+        {
+            Resources.Culture = new CultureInfo("en");
+            var tag = new TagModel(1, "a|b", RgbColor.Parse("#112233"), "first\nsecond");
+            string markdown = MarkdownExporter.Export(
+                "W",
+                [new ColumnModel(1, "C", string.Empty, 1, 1)],
+                [],
+                [(tag, 0L)],
+                ExportedAt);
+
+            StringAssert.Contains(markdown, "| a\\|b | #112233 | first second | 0 |");
+        }
+        finally
+        {
+            Resources.Culture = original;
+        }
+    }
+}
diff --git a/YKanBan.Tests/Search/SearchNodeComparer.cs b/YKanBan.Tests/Search/SearchNodeComparer.cs
new file mode 100644
index 0000000..652c3a4
--- /dev/null
+++ b/YKanBan.Tests/Search/SearchNodeComparer.cs
@@ -0,0 +1,69 @@
+using YKanBan.Search;
+
+namespace YKanBan.Tests.Search;
+
+/// <summary>
+/// Structural comparer for search AST nodes, used by assertions only. The AST
+/// records compare by reference for their Terms collections by design (see the
+/// YYC MARK note in SearchNodes.cs), so tests provide equality explicitly.
+/// </summary>
+public sealed class SearchNodeComparer : IEqualityComparer<SearchNode>
+{
+    /// <summary>
+    /// Gets the shared comparer instance.
+    /// </summary>
+    public static SearchNodeComparer Instance { get; } = new();
+
+    /// <summary>
+    /// Determines structural equality of two nodes.
+    /// </summary>
+    /// <param name="x">The first node.</param>
+    /// <param name="y">The second node.</param>
+    /// <returns><see langword="true"/> when the nodes are structurally equal.</returns>
+    public bool Equals(SearchNode? x, SearchNode? y)
+    {
+        if (ReferenceEquals(x, y))
+        {
+            return true;
+        }
+        if (x is null || y is null || x.GetType() != y.GetType())
+        {
+            return false;
+        }
+
+        return (x, y) switch
+        {
+            (SearchNode.Or a, SearchNode.Or b) => SequencesEqual(a.Terms, b.Terms),
+            (SearchNode.And a, SearchNode.And b) => SequencesEqual(a.Terms, b.Terms),
+            (SearchNode.FreeText a, SearchNode.FreeText b) => a.Text == b.Text,
+            (SearchNode.FieldText a, SearchNode.FieldText b) => a.Field == b.Field && a.Text == b.Text,
+            (SearchNode.TagName a, SearchNode.TagName b) => a.Name == b.Name,
+            (SearchNode.ColumnTitle a, SearchNode.ColumnTitle b) => a.Title == b.Title,
+            (SearchNode.CardId a, SearchNode.CardId b) => a.Id == b.Id,
+            _ => false,
+        };
+    }
+
+    /// <summary>
+    /// Not supported: the comparer exists for assertions, not hashing.
+    /// </summary>
+    /// <param name="obj">The node.</param>
+    /// <returns>Never returns.</returns>
+    public int GetHashCode(SearchNode obj) =>
+        throw new NotSupportedException("Structural comparison for assertions only; hashing is not supported.");
+
+    /// <summary>
+    /// Order-sensitive element-wise comparison.
+    /// </summary>
+    /// <param name="x">The first sequence.</param>
+    /// <param name="y">The second sequence.</param>
+    /// <returns><see langword="true"/> when both sequences match element-wise.</returns>
+    private static bool SequencesEqual(IReadOnlyList<SearchNode> x, IReadOnlyList<SearchNode> y)
+    {
+        if (x.Count != y.Count)
+        {
+            return false;
+        }
+        return x.Zip(y).All(pair => Instance.Equals(pair.First, pair.Second));
+    }
+}
diff --git a/YKanBan.Tests/Search/SearchQueryParserTests.cs b/YKanBan.Tests/Search/SearchQueryParserTests.cs
new file mode 100644
index 0000000..d17ce04
--- /dev/null
+++ b/YKanBan.Tests/Search/SearchQueryParserTests.cs
@@ -0,0 +1,292 @@
+using YKanBan.Search;
+
+namespace YKanBan.Tests.Search;
+
+/// <summary>
+/// Tests for the search parser: the full grammar branches plus the strict
+/// failure behavior (malformed input always throws
+/// <see cref="SearchSyntaxException"/>).
+/// </summary>
+[TestClass]
+public class SearchQueryParserTests
+{
+    /// <summary>
+    /// Parses and asserts structural equality against the expected AST.
+    /// </summary>
+    /// <param name="expected">The expected AST.</param>
+    /// <param name="text">The search text.</param>
+    private static void AssertParsesTo(SearchNode expected, string text)
+    {
+        SearchNode? parsed = SearchQueryParser.Parse(text);
+        Assert.IsNotNull(parsed, $"Blank-looking input: '{text}'.");
+        Assert.AreEqual(expected, parsed, SearchNodeComparer.Instance);
+    }
+
+    /// <summary>
+    /// Asserts that the input violates the grammar.
+    /// </summary>
+    /// <param name="text">The search text.</param>
+    private static void AssertSyntaxError(string text)
+    {
+        Assert.ThrowsExactly<SearchSyntaxException>(() => SearchQueryParser.Parse(text));
+    }
+
+    #region Successful parses
+
+    [TestMethod]
+    public void BlankInputReturnsNull()
+    {
+        Assert.IsNull(SearchQueryParser.Parse(string.Empty));
+        Assert.IsNull(SearchQueryParser.Parse("   "));
+    }
+
+    [TestMethod]
+    public void SingleWordIsFreeText()
+    {
+        AssertParsesTo(new SearchNode.FreeText("login"), "login");
+    }
+
+    [TestMethod]
+    public void AdjacentWordsAreImplicitAnd()
+    {
+        AssertParsesTo(
+            new SearchNode.And([new SearchNode.FreeText("fix"), new SearchNode.FreeText("login")]),
+            "fix login");
+    }
+
+    [TestMethod]
+    public void ExplicitAndKeyword()
+    {
+        AssertParsesTo(
+            new SearchNode.And([new SearchNode.FreeText("a"), new SearchNode.FreeText("b")]),
+            "a AND b");
+    }
+
+    [TestMethod]
+    public void OrKeyword()
+    {
+        AssertParsesTo(
+            new SearchNode.Or([new SearchNode.FreeText("a"), new SearchNode.FreeText("b")]),
+            "a OR b");
+    }
+
+    [TestMethod]
+    public void ImplicitAndBindsTighterThanOr()
+    {
+        AssertParsesTo(
+            new SearchNode.Or([
+                new SearchNode.And([new SearchNode.FreeText("a"), new SearchNode.FreeText("b")]),
+                new SearchNode.FreeText("c")]),
+            "a b OR c");
+    }
+
+    [TestMethod]
+    public void ParenthesesForceGrouping()
+    {
+        AssertParsesTo(
+            new SearchNode.And([
+                new SearchNode.FreeText("a"),
+                new SearchNode.Or([new SearchNode.FreeText("b"), new SearchNode.FreeText("c")])]),
+            "a (b OR c)");
+    }
+
+    [TestMethod]
+    public void NestedParenthesesCollapse()
+    {
+        AssertParsesTo(
+            new SearchNode.Or([new SearchNode.FreeText("a"), new SearchNode.FreeText("b")]),
+            "((a OR b))");
+    }
+
+    [TestMethod]
+    public void LowercaseKeywordsArePlainWords()
+    {
+        AssertParsesTo(
+            new SearchNode.And([
+                new SearchNode.FreeText("a"),
+                new SearchNode.FreeText("and"),
+                new SearchNode.FreeText("or"),
+                new SearchNode.FreeText("b")]),
+            "a and or b");
+    }
+
+    [TestMethod]
+    public void PhraseIsFreeText()
+    {
+        AssertParsesTo(new SearchNode.FreeText("hello world"), "\"hello world\"");
+    }
+
+    [TestMethod]
+    public void QuotedOperatorKeywordIsFreeText()
+    {
+        // A quoted AND/OR is literal text, not an operator.
+        AssertParsesTo(new SearchNode.FreeText("AND"), "\"AND\"");
+        AssertParsesTo(
+            new SearchNode.And([new SearchNode.FreeText("a"), new SearchNode.FreeText("OR")]),
+            "a \"OR\"");
+    }
+
+    [TestMethod]
+    public void PhraseSupportsEscapes()
+    {
+        // Escaped quote and escaped backslash.
+        AssertParsesTo(new SearchNode.FreeText("a\"b"), "\"a\\\"b\"");
+        AssertParsesTo(new SearchNode.FreeText("a\\b"), "\"a\\\\b\"");
+    }
+
+    [TestMethod]
+    public void TagQualifier()
+    {
+        AssertParsesTo(new SearchNode.TagName("bug"), "tag:bug");
+    }
+
+    [TestMethod]
+    public void TagQualifierAcceptsPhraseValue()
+    {
+        AssertParsesTo(new SearchNode.TagName("a b"), "tag:\"a b\"");
+    }
+
+    [TestMethod]
+    public void QualifierValueMayContainColons()
+    {
+        AssertParsesTo(new SearchNode.TagName("a:b"), "tag:a:b");
+    }
+
+    [TestMethod]
+    public void TitleAndContentQualifiers()
+    {
+        AssertParsesTo(new SearchNode.FieldText(SearchField.Title, "foo"), "title:foo");
+        AssertParsesTo(new SearchNode.FieldText(SearchField.Content, "foo"), "content:foo");
+    }
+
+    [TestMethod]
+    public void ColumnQualifierAcceptsPhraseValue()
+    {
+        AssertParsesTo(new SearchNode.ColumnTitle("To Do"), "column:\"To Do\"");
+    }
+
+    [TestMethod]
+    public void IdQualifierParsesNumber()
+    {
+        AssertParsesTo(new SearchNode.CardId(42), "id:42");
+        AssertParsesTo(new SearchNode.CardId(12), "id:\"12\"");
+    }
+
+    [TestMethod]
+    public void ColonWithoutKeyPrefixIsFreeText()
+    {
+        AssertParsesTo(new SearchNode.FreeText(":foo"), ":foo");
+    }
+
+    #endregion
+
+    #region Strict failures
+
+    [TestMethod]
+    public void UnterminatedPhraseIsAnError()
+    {
+        AssertSyntaxError("\"abc def");
+    }
+
+    [TestMethod]
+    public void TrailingUnterminatedPhraseIsAnError()
+    {
+        AssertSyntaxError("a \"abc");
+    }
+
+    [TestMethod]
+    public void InvalidEscapeInPhraseIsAnError()
+    {
+        // A bare backslash (not followed by a quote or another backslash) is invalid.
+        AssertSyntaxError("\"a\\b\"");
+    }
+
+    [TestMethod]
+    public void InvalidEscapeInQualifierPhraseIsAnError()
+    {
+        AssertSyntaxError("tag:\"a\\x\"");
+    }
+
+    [TestMethod]
+    public void UnbalancedOpenParenIsAnError()
+    {
+        AssertSyntaxError("(a");
+    }
+
+    [TestMethod]
+    public void StrayCloseParenIsAnError()
+    {
+        AssertSyntaxError("a)");
+    }
+
+    [TestMethod]
+    public void EmptyParensAreAnError()
+    {
+        AssertSyntaxError("()");
+    }
+
+    [TestMethod]
+    public void DanglingOrIsAnError()
+    {
+        AssertSyntaxError("a OR");
+    }
+
+    [TestMethod]
+    public void DanglingAndIsAnError()
+    {
+        AssertSyntaxError("a AND");
+    }
+
+    [TestMethod]
+    public void LeadingAndIsAnError()
+    {
+        AssertSyntaxError("AND a");
+    }
+
+    [TestMethod]
+    public void OperatorAtOperandPositionIsAnError()
+    {
+        AssertSyntaxError("a AND OR b");
+    }
+
+    [TestMethod]
+    public void QualifierWithoutValueIsAnError()
+    {
+        AssertSyntaxError("tag:");
+    }
+
+    [TestMethod]
+    public void SpacedQualifierValueIsAnError()
+    {
+        AssertSyntaxError("tag: bug");
+    }
+
+    [TestMethod]
+    public void NonNumericIdIsAnError()
+    {
+        AssertSyntaxError("id:12x");
+    }
+
+    [TestMethod]
+    public void UnknownQualifierKeyIsAnError()
+    {
+        AssertSyntaxError("foo:bar");
+    }
+
+    [TestMethod]
+    public void QualifierKeysAreCaseSensitive()
+    {
+        // "Tag" is a key-like prefix but not the known lowercase "tag", so it is unknown.
+        AssertSyntaxError("Tag:x");
+    }
+
+    [TestMethod]
+    public void SyntaxErrorCarriesDescriptiveMessage()
+    {
+        SearchSyntaxException exception = Assert.ThrowsExactly<SearchSyntaxException>(
+            () => SearchQueryParser.Parse("a AND"));
+        Assert.IsFalse(string.IsNullOrEmpty(exception.Message));
+    }
+
+    #endregion
+}
diff --git a/YKanBan.Tests/Search/SearchSqlCompilerTests.cs b/YKanBan.Tests/Search/SearchSqlCompilerTests.cs
new file mode 100644
index 0000000..836b4f5
--- /dev/null
+++ b/YKanBan.Tests/Search/SearchSqlCompilerTests.cs
@@ -0,0 +1,178 @@
+using Microsoft.Data.Sqlite;
+using YKanBan.Search;
+using YKanBan.Storage.Workspace;
+using YKanBan.Tests.TestUtilities;
+
+namespace YKanBan.Tests.Search;
+
+/// <summary>
+/// Tests for the query-to-SQL compiler, executed end-to-end against a fixture
+/// workspace database.
+/// </summary>
+[TestClass]
+public class SearchSqlCompilerTests
+{
+    /// <summary>
+    /// Owns a workspace with a fixed card/tag fixture.
+    /// </summary>
+    private sealed class SearchFixture : IDisposable
+    {
+        public TempDirectory Directory { get; } = new();
+
+        public SqliteConnection Connection { get; }
+
+        public SearchFixture()
+        {
+            WorkspaceInitializer.Initialize(Directory.FullPath);
+            using (SqliteConnection setup = SqliteTestHelper.OpenWorkspace(Directory.FullPath))
+            {
+                // Columns: 1 "To Do", 2 "In Progress".
+                SqliteTestHelper.Exec(setup, "INSERT INTO columns (title, description, created_at, updated_at) VALUES ('To Do', '', 1, 1);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO columns (title, description, created_at, updated_at) VALUES ('In Progress', '', 1, 1);");
+
+                // Cards: 1 titled, 2 title-or-content matches, 3 untitled, 4 wildcard chars.
+                SqliteTestHelper.Exec(setup, "INSERT INTO cards (column_id, title, content, created_at, updated_at) VALUES (1, 'Fix login', 'Cannot sign in', 1, 1);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO cards (column_id, title, content, created_at, updated_at) VALUES (2, 'Green apple', 'the apple pie', 1, 1);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO cards (column_id, title, content, created_at, updated_at) VALUES (2, '', 'banana split', 1, 1);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO cards (column_id, title, content, created_at, updated_at) VALUES (1, '100% done', 'a_b test', 1, 1);");
+
+                // Tags: mixed case names plus one containing spaces.
+                SqliteTestHelper.Exec(setup, "INSERT INTO tags (name, color, description) VALUES ('Bug', '#ff0000', '');");
+                SqliteTestHelper.Exec(setup, "INSERT INTO tags (name, color, description) VALUES ('UI', '#00ff00', '');");
+                SqliteTestHelper.Exec(setup, "INSERT INTO tags (name, color, description) VALUES ('release notes', '#0000ff', '');");
+                SqliteTestHelper.Exec(setup, "INSERT INTO card_tags (card_id, tag_id) VALUES (1, 1);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO card_tags (card_id, tag_id) VALUES (2, 2);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO card_tags (card_id, tag_id) VALUES (3, 2);");
+                SqliteTestHelper.Exec(setup, "INSERT INTO card_tags (card_id, tag_id) VALUES (4, 3);");
+            }
+
+            Connection = SqliteTestHelper.OpenWorkspace(Directory.FullPath);
+        }
+
+        /// <summary>
+        /// Runs a query and returns the ids of the matching cards.
+        /// </summary>
+        /// <param name="queryText">The search text.</param>
+        /// <returns>The matching card ids in id order.</returns>
+        public List<long> Search(string queryText)
+        {
+            SearchNode? query = SearchQueryParser.Parse(queryText);
+            Assert.IsNotNull(query, "Blank fixture queries are not expected here.");
+            CompiledSearch compiled = SearchSqlCompiler.Compile(query);
+
+            using var command = Connection.CreateCommand();
+            command.CommandText = $"SELECT c.id FROM cards c WHERE {compiled.Predicate} ORDER BY c.id ASC;";
+            foreach ((string name, object value) in compiled.Parameters)
+            {
+                command.Parameters.AddWithValue(name, value);
+            }
+            using var reader = command.ExecuteReader();
+
+            var ids = new List<long>();
+            while (reader.Read())
+            {
+                ids.Add(reader.GetInt64(0));
+            }
+            return ids;
+        }
+
+        public void Dispose()
+        {
+            Connection.Dispose();
+            Directory.Dispose();
+        }
+    }
+
+    [TestMethod]
+    public void FreeTextMatchesTitleOrContentCaseInsensitively()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 2L }, fixture.Search("apple"));
+        CollectionAssert.AreEqual(new[] { 1L }, fixture.Search("login"));
+        CollectionAssert.AreEqual(new[] { 2L }, fixture.Search("APPLE"));
+        CollectionAssert.AreEqual(new[] { 2L }, fixture.Search("pie"));
+    }
+
+    [TestMethod]
+    public void FieldQualifiersScopeToTheirField()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 2L }, fixture.Search("title:apple"));
+        CollectionAssert.AreEqual(new[] { 3L }, fixture.Search("content:banana"));
+        Assert.AreEqual(0, fixture.Search("title:pie").Count);
+    }
+
+    [TestMethod]
+    public void TagQualifierMatchesExactlyAndCaseInsensitively()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 1L }, fixture.Search("tag:bug"));
+        CollectionAssert.AreEqual(new[] { 2L, 3L }, fixture.Search("tag:ui"));
+        CollectionAssert.AreEqual(new[] { 4L }, fixture.Search("tag:\"release notes\""));
+
+        // Exact match: prefixes and substrings do not hit.
+        Assert.AreEqual(0, fixture.Search("tag:bu").Count);
+        Assert.AreEqual(0, fixture.Search("tag:release").Count);
+    }
+
+    [TestMethod]
+    public void ColumnQualifierMatchesColumnTitleExactly()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 1L, 4L }, fixture.Search("column:\"to do\""));
+        CollectionAssert.AreEqual(new[] { 2L, 3L }, fixture.Search("column:\"in progress\""));
+
+        // Not a prefix or substring match.
+        Assert.AreEqual(0, fixture.Search("column:To").Count);
+        Assert.AreEqual(0, fixture.Search("column:progress").Count);
+    }
+
+    [TestMethod]
+    public void IdQualifierMatchesExactly()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 3L }, fixture.Search("id:3"));
+        Assert.AreEqual(0, fixture.Search("id:99").Count);
+    }
+
+    [TestMethod]
+    public void AndOrCombination()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 2L, 3L }, fixture.Search("apple OR banana"));
+        CollectionAssert.AreEqual(new[] { 2L }, fixture.Search("apple AND id:2"));
+        Assert.AreEqual(0, fixture.Search("apple AND id:1").Count);
+    }
+
+    [TestMethod]
+    public void ParenthesizedGrouping()
+    {
+        using var fixture = new SearchFixture();
+
+        // (apple OR banana) hits cards 2,3; intersected with column "In Progress" (2,3).
+        CollectionAssert.AreEqual(new[] { 2L, 3L }, fixture.Search("(apple OR banana) AND column:\"In Progress\""));
+
+        // ...while column "To Do" holds cards 1,4 — the intersection is empty.
+        Assert.AreEqual(0, fixture.Search("(apple OR banana) AND column:\"To Do\"").Count);
+    }
+
+    [TestMethod]
+    public void LikeWildcardsInUserTextMatchLiterally()
+    {
+        using var fixture = new SearchFixture();
+        CollectionAssert.AreEqual(new[] { 4L }, fixture.Search("100%"));
+        CollectionAssert.AreEqual(new[] { 4L }, fixture.Search("a_b"));
+    }
+
+    [TestMethod]
+    public void CompiledPredicateUsesNamedParameters()
+    {
+        CompiledSearch compiled = SearchSqlCompiler.Compile(new SearchNode.FreeText("x"));
+
+        StringAssert.Contains(compiled.Predicate, "$p0");
+        StringAssert.Contains(compiled.Predicate, "$p1");
+        Assert.AreEqual(2, compiled.Parameters.Count);
+        Assert.AreEqual("%x%", compiled.Parameters[0].Value);
+        Assert.AreEqual("%x%", compiled.Parameters[1].Value);
+    }
+}
diff --git a/YKanBan/Assets/Locales/Resources.resx b/YKanBan/Assets/Locales/Resources.resx
index 8e09503..8a6f4c9 100644
--- a/YKanBan/Assets/Locales/Resources.resx
+++ b/YKanBan/Assets/Locales/Resources.resx
@@ -25,4 +25,34 @@
   <data name="PresetColumn_Done" xml:space="preserve">
     <value>Done</value>
   </data>
+  <data name="Export_FileNamePattern" xml:space="preserve">
+    <value>{0}-export-{1:yyyyMMdd-HHmmss}.md</value>
+  </data>
+  <data name="Export_TimePattern" xml:space="preserve">
+    <value>yyyy-MM-dd HH:mm:ss</value>
+  </data>
+  <data name="Export_CardTimePattern" xml:space="preserve">
+    <value>yyyy-MM-dd HH:mm</value>
+  </data>
+  <data name="Export_WorkspaceHeading" xml:space="preserve">
+    <value># {0}</value>
+  </data>
+  <data name="Export_ExportedAtLine" xml:space="preserve">
+    <value>&gt; Exported at: {0}</value>
+  </data>
+  <data name="Export_ColumnHeading" xml:space="preserve">
+    <value>## Column: {0} ({1})</value>
+  </data>
+  <data name="Export_TagsLine" xml:space="preserve">
+    <value>- Tags: {0}</value>
+  </data>
+  <data name="Export_CreatedModifiedLine" xml:space="preserve">
+    <value>- Created: {0}  Modified: {1}</value>
+  </data>
+  <data name="Export_TagsSummaryHeading" xml:space="preserve">
+    <value>## Tags</value>
+  </data>
+  <data name="Export_TagsTableHeader" xml:space="preserve">
+    <value>| Name | Color | Description | Usage |</value>
+  </data>
 </root>
diff --git a/YKanBan/Assets/Locales/Resources.zh-Hans.resx b/YKanBan/Assets/Locales/Resources.zh-Hans.resx
index 6eba378..162785e 100644
--- a/YKanBan/Assets/Locales/Resources.zh-Hans.resx
+++ b/YKanBan/Assets/Locales/Resources.zh-Hans.resx
@@ -25,4 +25,34 @@
   <data name="PresetColumn_Done" xml:space="preserve">
     <value>已完成</value>
   </data>
+  <data name="Export_FileNamePattern" xml:space="preserve">
+    <value>{0}-导出-{1:yyyyMMdd-HHmmss}.md</value>
+  </data>
+  <data name="Export_TimePattern" xml:space="preserve">
+    <value>yyyy年M月d日 HH:mm</value>
+  </data>
+  <data name="Export_CardTimePattern" xml:space="preserve">
+    <value>yyyy年M月d日 HH:mm</value>
+  </data>
+  <data name="Export_WorkspaceHeading" xml:space="preserve">
+    <value># {0}</value>
+  </data>
+  <data name="Export_ExportedAtLine" xml:space="preserve">
+    <value>&gt; 导出时间：{0}</value>
+  </data>
+  <data name="Export_ColumnHeading" xml:space="preserve">
+    <value>## 列：{0}（{1}）</value>
+  </data>
+  <data name="Export_TagsLine" xml:space="preserve">
+    <value>- 标签：{0}</value>
+  </data>
+  <data name="Export_CreatedModifiedLine" xml:space="preserve">
+    <value>- 创建：{0}　修改：{1}</value>
+  </data>
+  <data name="Export_TagsSummaryHeading" xml:space="preserve">
+    <value>## 标签总表</value>
+  </data>
+  <data name="Export_TagsTableHeader" xml:space="preserve">
+    <value>| 名称 | 颜色 | 描述 | 使用数 |</value>
+  </data>
 </root>
diff --git a/YKanBan/Export/MarkdownExporter.cs b/YKanBan/Export/MarkdownExporter.cs
new file mode 100644
index 0000000..e126545
--- /dev/null
+++ b/YKanBan/Export/MarkdownExporter.cs
@@ -0,0 +1,148 @@
+using System.Globalization;
+using System.Text;
+using YKanBan.Models;
+
+namespace YKanBan.Export;
+
+/// <summary>
+/// Generates the human-readable Markdown snapshot of a whole board. All
+/// template text and time formats come from the ResX resources in the current
+/// UI language; the export timestamp uses the per-language pattern, card
+/// timestamps use the shared minute-precision pattern. The output is meant for
+/// reading only — there is no re-import.
+/// </summary>
+public static class MarkdownExporter
+{
+    /// <summary>
+    /// Renders the whole board to Markdown.
+    /// </summary>
+    /// <param name="workspaceName">Workspace display name (the folder name).</param>
+    /// <param name="columns">All columns, in id order; empty columns are kept.</param>
+    /// <param name="cards">All cards; grouped by column, id order within a column.</param>
+    /// <param name="tags">All tags with usage counts, in id order.</param>
+    /// <param name="exportedAt">Export moment in local time.</param>
+    /// <returns>The Markdown document.</returns>
+    public static string Export(
+        string workspaceName,
+        IReadOnlyList<ColumnModel> columns,
+        IReadOnlyList<CardModel> cards,
+        IReadOnlyList<(TagModel Tag, long UsageCount)> tags,
+        DateTimeOffset exportedAt)
+    {
+        CultureInfo culture = Resources.Culture ?? CultureInfo.CurrentUICulture;
+
+        var builder = new StringBuilder();
+
+        // Header: workspace name and the export timestamp in the UI language's format.
+        builder.AppendLine(string.Format(culture, Resources.Export_WorkspaceHeading, workspaceName));
+        builder.AppendLine(string.Format(
+            culture, Resources.Export_ExportedAtLine, exportedAt.ToString(Resources.Export_TimePattern, culture)));
+        builder.AppendLine();
+
+        // Board body: every column in id order, empty ones included.
+        foreach (ColumnModel column in columns)
+        {
+            IReadOnlyList<CardModel> cardsInColumn = cards.Where(card => card.ColumnId == column.Id).ToArray();
+            builder.AppendLine(string.Format(culture, Resources.Export_ColumnHeading, column.Title, cardsInColumn.Count));
+            if (column.Description.Length > 0)
+            {
+                builder.AppendLine(SingleLine(column.Description));
+            }
+            builder.AppendLine();
+
+            foreach (CardModel card in cardsInColumn)
+            {
+                AppendCard(builder, card, culture);
+            }
+        }
+
+        AppendTagTable(builder, tags, culture);
+        return builder.ToString();
+    }
+
+    /// <summary>
+    /// Builds the default export file name in the UI language, for example
+    /// "MyRepo-export-20260930-142537.md".
+    /// </summary>
+    /// <param name="workspaceName">Workspace display name.</param>
+    /// <param name="exportedAt">Export moment.</param>
+    /// <returns>The file name.</returns>
+    public static string BuildFileName(string workspaceName, DateTimeOffset exportedAt)
+    {
+        CultureInfo culture = Resources.Culture ?? CultureInfo.CurrentUICulture;
+        return string.Format(culture, Resources.Export_FileNamePattern, workspaceName, exportedAt);
+    }
+
+    /// <summary>
+    /// Appends one card block: heading, body, tags line and timestamps line.
+    /// </summary>
+    /// <param name="builder">The output builder.</param>
+    /// <param name="card">The card to render.</param>
+    /// <param name="culture">The UI culture.</param>
+    private static void AppendCard(StringBuilder builder, CardModel card, CultureInfo culture)
+    {
+        // Heading: "#id title"; an untitled card shows only the id.
+        builder.AppendLine(card.Title.Length == 0
+            ? $"### #{card.Id}"
+            : $"### #{card.Id} {SingleLine(card.Title)}");
+
+        if (card.Content.Length > 0)
+        {
+            builder.AppendLine(card.Content);
+        }
+
+        if (card.Tags.Count > 0)
+        {
+            string names = string.Join(" ", card.Tags.Select(tag => $"`{tag.Name}`"));
+            builder.AppendLine(string.Format(culture, Resources.Export_TagsLine, names));
+        }
+
+        string created = ToLocalTime(card.CreatedAt).ToString(Resources.Export_CardTimePattern, culture);
+        string updated = ToLocalTime(card.UpdatedAt).ToString(Resources.Export_CardTimePattern, culture);
+        builder.AppendLine(string.Format(culture, Resources.Export_CreatedModifiedLine, created, updated));
+        builder.AppendLine();
+    }
+
+    /// <summary>
+    /// Appends the tag summary table with usage counts; tags with zero usage are included.
+    /// </summary>
+    /// <param name="builder">The output builder.</param>
+    /// <param name="tags">The tags and their usage counts.</param>
+    /// <param name="culture">The UI culture.</param>
+    private static void AppendTagTable(
+        StringBuilder builder,
+        IReadOnlyList<(TagModel Tag, long UsageCount)> tags,
+        CultureInfo culture)
+    {
+        builder.AppendLine(Resources.Export_TagsSummaryHeading);
+        builder.AppendLine(Resources.Export_TagsTableHeader);
+        builder.AppendLine("|---|---|---|---|");
+        foreach ((TagModel tag, long usage) in tags)
+        {
+            // '|' and line breaks would break the table row; neutralize them.
+            builder.AppendLine($"| {Cell(tag.Name)} | {tag.Color} | {Cell(tag.Description)} | {usage} |");
+        }
+    }
+
+    /// <summary>
+    /// Formats a table cell: single line with escaped pipe characters.
+    /// </summary>
+    /// <param name="text">The raw cell text.</param>
+    /// <returns>The escaped single-line text.</returns>
+    private static string Cell(string text) => SingleLine(text).Replace("|", "\\|");
+
+    /// <summary>
+    /// Collapses line breaks so the text stays on one Markdown line.
+    /// </summary>
+    /// <param name="text">The raw text.</param>
+    /// <returns>The text without line breaks.</returns>
+    private static string SingleLine(string text) => text.Replace("\r", string.Empty).Replace("\n", " ");
+
+    /// <summary>
+    /// Converts stored Unix seconds to a local-time <see cref="DateTimeOffset"/>.
+    /// </summary>
+    /// <param name="unixSeconds">The stored timestamp.</param>
+    /// <returns>The local time.</returns>
+    private static DateTimeOffset ToLocalTime(long unixSeconds) =>
+        DateTimeOffset.FromUnixTimeSeconds(unixSeconds).ToLocalTime();
+}
diff --git a/YKanBan/Search/SearchLexer.cs b/YKanBan/Search/SearchLexer.cs
new file mode 100644
index 0000000..0da560b
--- /dev/null
+++ b/YKanBan/Search/SearchLexer.cs
@@ -0,0 +1,104 @@
+using Pidgin;
+
+namespace YKanBan.Search;
+
+/// <summary>
+/// Lexical tokens of the search syntax.
+/// </summary>
+public abstract record SearchToken
+{
+    /// <summary>
+    /// A bare word; qualifier detection happens later from the text.
+    /// </summary>
+    /// <param name="Text">The word text.</param>
+    public sealed record Word(string Text) : SearchToken;
+
+    /// <summary>
+    /// A quoted phrase with its escapes already resolved.
+    /// </summary>
+    /// <param name="Text">The unescaped phrase content.</param>
+    public sealed record Phrase(string Text) : SearchToken;
+
+    /// <summary>
+    /// An opening parenthesis.
+    /// </summary>
+    public sealed record OpenParen : SearchToken;
+
+    /// <summary>
+    /// A closing parenthesis.
+    /// </summary>
+    public sealed record CloseParen : SearchToken;
+}
+
+/// <summary>
+/// Pidgin-based lexer for the search syntax. Splits the input into words,
+/// quoted phrases and parentheses; whitespace only separates tokens.
+///
+/// The lexer is strict: inside a phrase only <c>\"</c> and <c>\\</c> are valid
+/// escapes, a bare backslash or an unknown escape is rejected, and an
+/// unterminated phrase is rejected. Any input not covered by the token rules
+/// is rejected rather than silently ignored.
+/// </summary>
+public static class SearchLexer
+{
+    /// <summary>A valid phrase escape sequence: <c>\"</c> or <c>\\</c>.</summary>
+    private static readonly Parser<char, char> PhraseEscape =
+        Parser.Try(Parser.String("\\\"").Select(_ => '"'))
+            .Or(Parser.String("\\\\").Select(_ => '\\'));
+
+    /// <summary>One phrase character: an escape or any character except a quote or backslash.</summary>
+    private static readonly Parser<char, char> PhraseCharacter =
+        PhraseEscape.Or(Parser<char>.Token(character => character != '"' && character != '\\'));
+
+    private static readonly Parser<char, SearchToken> PhraseParser =
+        from openingQuote in Parser.Char('"')
+        from content in PhraseCharacter.Many()
+        from closingQuote in Parser.Char('"')
+        select (SearchToken)new SearchToken.Phrase(new string(content.ToArray()));
+
+    private static readonly Parser<char, SearchToken> OpenParenParser =
+        Parser.Char('(').Select(_ => (SearchToken)new SearchToken.OpenParen());
+
+    private static readonly Parser<char, SearchToken> CloseParenParser =
+        Parser.Char(')').Select(_ => (SearchToken)new SearchToken.CloseParen());
+
+    private static readonly Parser<char, SearchToken> WordParser =
+        Parser<char>.Token(character =>
+                !char.IsWhiteSpace(character) && character != '(' && character != ')' && character != '"')
+            .AtLeastOnce()
+            .Select(chars => (SearchToken)new SearchToken.Word(new string(chars.ToArray())));
+
+    private static readonly Parser<char, SearchToken> TokenParser =
+        Parser.SkipWhitespaces.Then(
+            PhraseParser
+                .Or(OpenParenParser)
+                .Or(CloseParenParser)
+                .Or(WordParser));
+
+    // Try(...) keeps a failed token non-consuming so Many stops gracefully; End
+    // then rejects any leftover input, which turns dangling quotes or invalid
+    // escapes into hard failures instead of silently dropped text.
+    private static readonly Parser<char, IReadOnlyList<SearchToken>> Lexer =
+        Parser.Try(TokenParser).Many()
+            .Before(Parser.SkipWhitespaces)
+            .Before(Parser<char>.End)
+            .Select(tokens => (IReadOnlyList<SearchToken>)tokens.ToList());
+
+    /// <summary>
+    /// Tokenizes the search text.
+    /// </summary>
+    /// <param name="input">The raw search text.</param>
+    /// <returns>The token stream.</returns>
+    /// <exception cref="SearchSyntaxException">The text contains an invalid escape or an unterminated phrase.</exception>
+    public static IReadOnlyList<SearchToken> Tokenize(string input)
+    {
+        try
+        {
+            return Lexer.ParseOrThrow(input);
+        }
+        catch (ParseException exception)
+        {
+            throw new SearchSyntaxException(exception.Message, exception);
+        }
+    }
+}
diff --git a/YKanBan/Search/SearchNodes.cs b/YKanBan/Search/SearchNodes.cs
new file mode 100644
index 0000000..5609615
--- /dev/null
+++ b/YKanBan/Search/SearchNodes.cs
@@ -0,0 +1,69 @@
+// YYC MARK: C# records synthesize equality by running EqualityComparer<T>.Default
+// on each positional property. For collection properties (IReadOnlyList<>,
+// ImmutableArray<>, arrays, ...) that comparer falls back to REFERENCE equality,
+// so the Or/And nodes below deliberately do NOT compare structurally. The only
+// structural comparison lives in the test helper (SearchNodeComparer); production
+// code never compares AST nodes.
+namespace YKanBan.Search;
+
+/// <summary>
+/// Root of the search query AST. Search keywords are fixed English grammar
+/// and are never localized.
+/// </summary>
+public abstract record SearchNode
+{
+    /// <summary>
+    /// Disjunction: matches when any term matches (OR).
+    /// </summary>
+    /// <param name="Terms">The alternative terms.</param>
+    public sealed record Or(IReadOnlyList<SearchNode> Terms) : SearchNode;
+
+    /// <summary>
+    /// Conjunction: matches when all terms match (explicit AND or adjacency).
+    /// </summary>
+    /// <param name="Terms">The required terms.</param>
+    public sealed record And(IReadOnlyList<SearchNode> Terms) : SearchNode;
+
+    /// <summary>
+    /// A bare word or phrase matched as a substring of the title OR content.
+    /// </summary>
+    /// <param name="Text">The search text.</param>
+    public sealed record FreeText(string Text) : SearchNode;
+
+    /// <summary>
+    /// A substring match scoped to one card field.
+    /// </summary>
+    /// <param name="Field">The card field to match against.</param>
+    /// <param name="Text">The search text.</param>
+    public sealed record FieldText(SearchField Field, string Text) : SearchNode;
+
+    /// <summary>
+    /// An exact, case-insensitive tag name match.
+    /// </summary>
+    /// <param name="Name">The tag name to match.</param>
+    public sealed record TagName(string Name) : SearchNode;
+
+    /// <summary>
+    /// An exact, case-insensitive column title match.
+    /// </summary>
+    /// <param name="Title">The column title to match.</param>
+    public sealed record ColumnTitle(string Title) : SearchNode;
+
+    /// <summary>
+    /// An exact card id match.
+    /// </summary>
+    /// <param name="Id">The card id to match.</param>
+    public sealed record CardId(long Id) : SearchNode;
+}
+
+/// <summary>
+/// The card fields a <see cref="SearchNode.FieldText"/> atom can scope to.
+/// </summary>
+public enum SearchField
+{
+    /// <summary>The card title.</summary>
+    Title,
+
+    /// <summary>The card content body.</summary>
+    Content,
+}
diff --git a/YKanBan/Search/SearchQueryParser.cs b/YKanBan/Search/SearchQueryParser.cs
new file mode 100644
index 0000000..47b8054
--- /dev/null
+++ b/YKanBan/Search/SearchQueryParser.cs
@@ -0,0 +1,257 @@
+using System.Globalization;
+using Pidgin;
+
+namespace YKanBan.Search;
+
+/// <summary>
+/// Parses the search grammar with Pidgin, strictly: any malformed input
+/// (unbalanced parentheses, dangling or consecutive operators, unterminated
+/// phrases, invalid escapes, qualifier keys without a value or with an
+/// unknown/non-numeric id value) raises <see cref="SearchSyntaxException"/>.
+///
+/// Only uppercase AND/OR are operators (lowercase ones are plain words); AND —
+/// explicit or by adjacency — binds tighter than OR; parentheses group. Blank
+/// input returns null and means "show everything". Search keywords and
+/// qualifier keys are fixed English grammar and are never localized.
+/// </summary>
+public static class SearchQueryParser
+{
+    /// <summary>The recognized qualifier keys; add new entries here to extend the grammar.</summary>
+    private static readonly (string Key, QualifierKind Kind)[] QualifierTable =
+    [
+        ("tag", QualifierKind.Tag),
+        ("title", QualifierKind.Title),
+        ("content", QualifierKind.Content),
+        ("id", QualifierKind.Id),
+        ("column", QualifierKind.Column),
+    ];
+
+    #region Token primitives
+
+    private static readonly Parser<SearchToken, SearchToken> AndKeyword =
+        Parser<SearchToken>.Token(token => token is SearchToken.Word { Text: "AND" });
+
+    private static readonly Parser<SearchToken, SearchToken> OrKeyword =
+        Parser<SearchToken>.Token(token => token is SearchToken.Word { Text: "OR" });
+
+    private static readonly Parser<SearchToken, SearchToken> OpenParen =
+        Parser<SearchToken>.Token(token => token is SearchToken.OpenParen);
+
+    private static readonly Parser<SearchToken, SearchToken> CloseParen =
+        Parser<SearchToken>.Token(token => token is SearchToken.CloseParen);
+
+    /// <summary>The text of a phrase token.</summary>
+    private static readonly Parser<SearchToken, string> PhraseText =
+        Parser<SearchToken>.Token(token => token is SearchToken.Phrase)
+            .Select(token => ((SearchToken.Phrase)token).Text);
+
+    /// <summary>The text of a word token, operator keywords excluded (they never start an operand).</summary>
+    private static readonly Parser<SearchToken, string> NonKeywordWord =
+        Parser<SearchToken>.Token(token => token is SearchToken.Word word && word.Text != "AND" && word.Text != "OR")
+            .Select(token => ((SearchToken.Word)token).Text);
+
+    #endregion
+
+    #region Grammar
+
+    // A bare phrase matched against title or content.
+    private static readonly Parser<SearchToken, SearchNode> FreePhrase =
+        PhraseText.Select(text => (SearchNode)new SearchNode.FreeText(text));
+
+    // A bare word: a "key:value" qualifier when the prefix matches, otherwise free text.
+    private static readonly Parser<SearchToken, SearchNode> WordAtom =
+        NonKeywordWord.Bind(QualifierFrom);
+
+    // '(' orExpr ')' — the deferred method call breaks the
+    // Parenthesized → OrExpr → AndExpr → Operand → Parenthesized recursion cycle.
+    private static readonly Parser<SearchToken, SearchNode> Parenthesized =
+        OpenParen.Then(_ => OrExprDeferred()).Before(CloseParen);
+
+    // operand := '(' orExpr ')' | qualifier-or-word | phrase
+    private static readonly Parser<SearchToken, SearchNode> Operand =
+        Parenthesized.Or(WordAtom).Or(FreePhrase);
+
+    // One andExpr operand together with its optional explicit AND keyword.
+    private static readonly Parser<SearchToken, SearchNode> AndPrefixedOperand =
+        AndKeyword.Optional().Then(Operand);
+
+    // andExpr := operand (AND? operand)*
+    private static readonly Parser<SearchToken, SearchNode> AndExpr =
+        Operand.Bind(first =>
+            AndPrefixedOperand.Many()
+                .Select(rest => FoldAnd(first, rest)));
+
+    // orExpr := andExpr (OR andExpr)+
+    private static readonly Parser<SearchToken, SearchNode> OrExpr =
+        AndExpr.SeparatedAtLeastOnce(OrKeyword)
+            .Select(FoldOr);
+
+    // The full query: orExpr consuming the whole token stream.
+    private static readonly Parser<SearchToken, SearchNode> Query =
+        OrExpr.Before(Parser<SearchToken>.End);
+
+    /// <summary>Defers the OrExpr field read to parse time (recursion cycle breaker).</summary>
+    /// <returns>The OR-expression parser.</returns>
+    private static Parser<SearchToken, SearchNode> OrExprDeferred() => OrExpr;
+
+    #endregion
+
+    /// <summary>
+    /// Parses a search box text.
+    /// </summary>
+    /// <param name="text">The raw search text.</param>
+    /// <returns>The parsed query, or <see langword="null"/> for blank input (meaning: show everything).</returns>
+    /// <exception cref="SearchSyntaxException">The text violates the grammar.</exception>
+    public static SearchNode? Parse(string text)
+    {
+        if (string.IsNullOrWhiteSpace(text))
+        {
+            return null;
+        }
+
+        try
+        {
+            List<SearchToken> tokens = SearchLexer.Tokenize(text).ToList();
+            return Query.ParseOrThrow(tokens);
+        }
+        catch (ParseException exception)
+        {
+            throw new SearchSyntaxException(exception.Message, exception);
+        }
+    }
+
+    #region Atom construction
+
+    /// <summary>
+    /// Builds the parser for one word operand: a qualifier atom, an unknown-key
+    /// failure, or free text.
+    /// </summary>
+    /// <param name="word">The word token text.</param>
+    /// <returns>The parser producing the operand node.</returns>
+    private static Parser<SearchToken, SearchNode> QualifierFrom(string word)
+    {
+        int colonIndex = word.IndexOf(':');
+
+        // Only a leading run of letters/digits/underscores before ':' can be a qualifier key.
+        if (colonIndex <= 0 || !IsQualifierKeyCandidate(word.AsSpan(0, colonIndex)))
+        {
+            return Parser<SearchToken>.Return<SearchNode>(new SearchNode.FreeText(word));
+        }
+
+        string key = word[..colonIndex];
+        string value = word[(colonIndex + 1)..];
+
+        if (!TryLookupQualifier(key, out QualifierKind kind))
+        {
+            return Parser<SearchToken>.Fail<SearchNode>($"Unknown qualifier key '{key}'.");
+        }
+
+        // "key:" without an inline value takes the value from the following phrase, e.g. tag:"a b".
+        return value.Length == 0
+            ? PhraseText.Bind(text => AtomParser(kind, text))
+            : AtomParser(kind, value);
+    }
+
+    /// <summary>
+    /// Builds the atom parser for a qualifier value, failing strictly on a non-numeric id.
+    /// </summary>
+    /// <param name="kind">The qualifier kind.</param>
+    /// <param name="value">The qualifier value.</param>
+    /// <returns>The parser producing the atom node.</returns>
+    private static Parser<SearchToken, SearchNode> AtomParser(QualifierKind kind, string value) =>
+        AtomFrom(kind, value) is { } atom
+            ? Parser<SearchToken>.Return(atom)
+            : Parser<SearchToken>.Fail<SearchNode>($"The id qualifier expects an integer, got '{value}'.");
+
+    /// <summary>
+    /// Builds the atom for a qualifier value; null when an id value is not an integer.
+    /// </summary>
+    /// <param name="kind">The qualifier kind.</param>
+    /// <param name="value">The qualifier value.</param>
+    /// <returns>The atom node, or <see langword="null"/> for an invalid id value.</returns>
+    private static SearchNode? AtomFrom(QualifierKind kind, string value) => kind switch
+    {
+        QualifierKind.Tag => new SearchNode.TagName(value),
+        QualifierKind.Title => new SearchNode.FieldText(SearchField.Title, value),
+        QualifierKind.Content => new SearchNode.FieldText(SearchField.Content, value),
+        QualifierKind.Column => new SearchNode.ColumnTitle(value),
+        QualifierKind.Id when long.TryParse(value, NumberStyles.Integer, CultureInfo.InvariantCulture, out long id) =>
+            new SearchNode.CardId(id),
+        _ => null,
+    };
+
+    /// <summary>
+    /// Looks up a qualifier key in the extension table with ordinal (case-sensitive) matching.
+    /// </summary>
+    /// <param name="key">The key text before the colon.</param>
+    /// <param name="kind">The matching qualifier kind.</param>
+    /// <returns><see langword="true"/> when the key is known.</returns>
+    private static bool TryLookupQualifier(string key, out QualifierKind kind)
+    {
+        foreach ((string candidate, QualifierKind candidateKind) in QualifierTable)
+        {
+            if (string.Equals(candidate, key, StringComparison.Ordinal))
+            {
+                kind = candidateKind;
+                return true;
+            }
+        }
+
+        kind = default;
+        return false;
+    }
+
+    /// <summary>
+    /// Returns whether the text looks like a qualifier key: a non-empty run of
+    /// letters, digits or underscores.
+    /// </summary>
+    /// <param name="text">The candidate key text.</param>
+    /// <returns><see langword="true"/> when every character is a word character.</returns>
+    private static bool IsQualifierKeyCandidate(ReadOnlySpan<char> text)
+    {
+        foreach (char character in text)
+        {
+            if (!(char.IsLetterOrDigit(character) || character == '_'))
+            {
+                return false;
+            }
+        }
+
+        return true;
+    }
+
+    /// <summary>
+    /// Collapses an AND chain into a single node.
+    /// </summary>
+    /// <param name="first">The first operand.</param>
+    /// <param name="rest">The remaining operands.</param>
+    /// <returns>The single operand, or an <see cref="SearchNode.And"/> node.</returns>
+    private static SearchNode FoldAnd(SearchNode first, IEnumerable<SearchNode> rest)
+    {
+        List<SearchNode> terms = [first, .. rest];
+        return terms.Count == 1 ? terms[0] : new SearchNode.And(terms);
+    }
+
+    /// <summary>
+    /// Collapses an OR chain into a single node.
+    /// </summary>
+    /// <param name="terms">The operands.</param>
+    /// <returns>The single operand, or a <see cref="SearchNode.Or"/> node.</returns>
+    private static SearchNode FoldOr(IEnumerable<SearchNode> terms)
+    {
+        List<SearchNode> list = terms.ToList();
+        return list.Count == 1 ? list[0] : new SearchNode.Or(list);
+    }
+
+    #endregion
+
+    /// <summary>The recognized qualifier kinds.</summary>
+    private enum QualifierKind
+    {
+        Tag,
+        Title,
+        Content,
+        Id,
+        Column,
+    }
+}
diff --git a/YKanBan/Search/SearchSqlCompiler.cs b/YKanBan/Search/SearchSqlCompiler.cs
new file mode 100644
index 0000000..95530fe
--- /dev/null
+++ b/YKanBan/Search/SearchSqlCompiler.cs
@@ -0,0 +1,107 @@
+namespace YKanBan.Search;
+
+/// <summary>
+/// A compiled search predicate: a SQL fragment over the "c" (cards) alias plus
+/// its bound parameters.
+/// </summary>
+/// <param name="Predicate">SQL boolean expression, for <c>SELECT ... FROM cards c WHERE {Predicate}</c>.</param>
+/// <param name="Parameters">Named parameters referenced by the predicate, in order.</param>
+public sealed record CompiledSearch(string Predicate, IReadOnlyList<(string Name, object Value)> Parameters);
+
+/// <summary>
+/// Compiles a parsed search AST into a parameterized SQL predicate. Matching is
+/// case-insensitive with SQLite's built-in semantics: LIKE's default case
+/// folding for substring matches plus COLLATE NOCASE for exact matches. LIKE
+/// wildcards and the escape character in user text are escaped so they match
+/// literally.
+/// </summary>
+public static class SearchSqlCompiler
+{
+    /// <summary>
+    /// Compiles the query into a SQL predicate.
+    /// </summary>
+    /// <param name="query">The parsed search AST.</param>
+    /// <returns>The compiled predicate and its parameters.</returns>
+    public static CompiledSearch Compile(SearchNode query)
+    {
+        var parameters = new List<(string Name, object Value)>();
+        string predicate = Build(query, parameters);
+        return new CompiledSearch(predicate, parameters);
+    }
+
+    /// <summary>
+    /// Recursively builds the SQL expression for one node.
+    /// </summary>
+    /// <param name="node">The AST node to compile.</param>
+    /// <param name="parameters">The parameter accumulator.</param>
+    /// <returns>The SQL boolean expression for the node.</returns>
+    private static string Build(SearchNode node, List<(string Name, object Value)> parameters)
+    {
+        switch (node)
+        {
+            case SearchNode.And and:
+                return "(" + string.Join(" AND ", and.Terms.Select(term => Build(term, parameters))) + ")";
+
+            case SearchNode.Or or:
+                return "(" + string.Join(" OR ", or.Terms.Select(term => Build(term, parameters))) + ")";
+
+            case SearchNode.FreeText freeText:
+            {
+                // A bare word matches a substring of the title OR the content.
+                string titleParam = AddParameter(parameters, "%" + EscapeLike(freeText.Text) + "%");
+                string contentParam = AddParameter(parameters, "%" + EscapeLike(freeText.Text) + "%");
+                return $"(c.title LIKE {titleParam} ESCAPE '\\' OR c.content LIKE {contentParam} ESCAPE '\\')";
+            }
+
+            case SearchNode.FieldText field:
+            {
+                string column = field.Field == SearchField.Title ? "c.title" : "c.content";
+                string param = AddParameter(parameters, "%" + EscapeLike(field.Text) + "%");
+                return $"{column} LIKE {param} ESCAPE '\\'";
+            }
+
+            case SearchNode.TagName tag:
+            {
+                string param = AddParameter(parameters, tag.Name);
+                return "EXISTS (SELECT 1 FROM card_tags ct JOIN tags t ON t.id = ct.tag_id " +
+                       $"WHERE ct.card_id = c.id AND t.name COLLATE NOCASE = {param})";
+            }
+
+            case SearchNode.ColumnTitle column:
+            {
+                string param = AddParameter(parameters, column.Title);
+                return $"c.column_id IN (SELECT id FROM columns WHERE title COLLATE NOCASE = {param})";
+            }
+
+            case SearchNode.CardId id:
+            {
+                string param = AddParameter(parameters, id.Id);
+                return $"c.id = {param}";
+            }
+
+            default:
+                throw new InvalidOperationException($"Unsupported search node: {node.GetType().Name}");
+        }
+    }
+
+    /// <summary>
+    /// Binds a value under a fresh name and returns the SQL placeholder.
+    /// </summary>
+    /// <param name="parameters">The parameter accumulator.</param>
+    /// <param name="value">The value to bind.</param>
+    /// <returns>The placeholder text.</returns>
+    private static string AddParameter(List<(string Name, object Value)> parameters, object value)
+    {
+        string name = "$p" + parameters.Count.ToString(System.Globalization.CultureInfo.InvariantCulture);
+        parameters.Add((name, value));
+        return name;
+    }
+
+    /// <summary>
+    /// Escapes LIKE wildcards and the escape character inside user text.
+    /// </summary>
+    /// <param name="text">The raw user text.</param>
+    /// <returns>The text with <c>\</c>, <c>%</c> and <c>_</c> escaped.</returns>
+    private static string EscapeLike(string text) =>
+        text.Replace("\\", "\\\\").Replace("%", "\\%").Replace("_", "\\_");
+}
diff --git a/YKanBan/Search/SearchSyntaxException.cs b/YKanBan/Search/SearchSyntaxException.cs
new file mode 100644
index 0000000..483d89a
--- /dev/null
+++ b/YKanBan/Search/SearchSyntaxException.cs
@@ -0,0 +1,20 @@
+namespace YKanBan.Search;
+
+/// <summary>
+/// Thrown when the search text violates the search grammar. Carries the
+/// underlying parser's descriptive English message; the UI surfaces it as a
+/// search error instead of coupling to Pidgin types. The message is never
+/// localized.
+/// </summary>
+public sealed class SearchSyntaxException : Exception
+{
+    /// <summary>
+    /// Initializes the exception with a descriptive English message.
+    /// </summary>
+    /// <param name="message">The English diagnostic message.</param>
+    /// <param name="innerException">The underlying parse failure.</param>
+    public SearchSyntaxException(string message, Exception innerException)
+        : base(message, innerException)
+    {
+    }
+}