diff --git a/pkg/channels/slack/parse_markdown_to_mrkdwn.go b/pkg/channels/slack/parse_markdown_to_mrkdwn.go
new file mode 100644
index 000000000..514cf597d
--- /dev/null
+++ b/pkg/channels/slack/parse_markdown_to_mrkdwn.go
@@ -0,0 +1,177 @@
+package slack
+
+import (
+	"fmt"
+	"regexp"
+	"strings"
+)
+
+var (
+	// Patterns are applied in order; code blocks extracted first to protect verbatim content.
+	reMrkdwnCodeBlock  = regexp.MustCompile("(?s)```[\\w]*\\n?([\\s\\S]*?)```")
+	reMrkdwnInlineCode = regexp.MustCompile("`([^`\n]+)`")
+	reMrkdwnHeading    = regexp.MustCompile(`(?m)^#{1,6}\s+(.+)$`)
+	reMrkdwnBoldStar   = regexp.MustCompile(`\*\*(.+?)\*\*`)
+	reMrkdwnBoldUnder  = regexp.MustCompile(`__(.+?)__`)
+	reMrkdwnStrike     = regexp.MustCompile(`~~(.+?)~~`)
+	reMrkdwnImage      = regexp.MustCompile(`!\[([^\]]*)\]\(([^)]+)\)`)
+	reMrkdwnLink       = regexp.MustCompile(`\[([^\]]+)\]\(([^)]+)\)`)
+	reMrkdwnHR         = regexp.MustCompile(`(?m)^[-*_]{3,}\s*$`)
+	reMrkdwnULItem     = regexp.MustCompile(`(?m)^(\s*)[-*+]\s+`)
+	// Table: line starting with optional whitespace, pipe, content, pipe.
+	reMrkdwnTableRow = regexp.MustCompile(`(?m)^\|(.+)\|$`)
+	// Separator row: | --- | --- | or | :---: | etc.
+	reMrkdwnTableSep = regexp.MustCompile(`(?m)^\|[\s|:-]+\|$`)
+)
+
+// markdownToSlackMrkdwn converts standard Markdown (as produced by LLMs) into
+// Slack's mrkdwn format.
+//
+// Conversion rules:
+//   - Fenced code blocks and inline code are preserved unchanged.
+//   - Markdown tables are converted to preformatted code blocks.
+//   - Headings (# … ######) become *bold* text.
+//   - **bold** and __bold__ become *bold*.
+//   - ~~strikethrough~~ becomes ~strikethrough~.
+//   - [text](url) becomes <url|text>.
+//   - ![alt](url) becomes <url|alt>.
+//   - Horizontal rules (---) are removed.
+//   - Unordered list markers (-, *, +) become bullet characters (•).
+//   - Italic (_text_), blockquotes (> text), and code pass through unchanged.
+func markdownToSlackMrkdwn(text string) string {
+	if text == "" {
+		return ""
+	}
+
+	// 1. Extract code blocks and inline code to protect them from conversion.
+	var codeBlocks []string
+	text = reMrkdwnCodeBlock.ReplaceAllStringFunc(text, func(match string) string {
+		idx := len(codeBlocks)
+		codeBlocks = append(codeBlocks, match)
+		return codePlaceholder(idx)
+	})
+
+	var inlineCodes []string
+	text = reMrkdwnInlineCode.ReplaceAllStringFunc(text, func(match string) string {
+		idx := len(inlineCodes)
+		inlineCodes = append(inlineCodes, match)
+		return inlineCodePlaceholder(idx)
+	})
+
+	// 2. Convert tables to preformatted code blocks.
+	text = convertTables(text)
+
+	// 3. Headings → bold. Strip any bold markers from heading content first
+	//    since the heading itself is rendered as bold.
+	text = reMrkdwnHeading.ReplaceAllStringFunc(text, func(match string) string {
+		sub := reMrkdwnHeading.FindStringSubmatch(match)
+		if len(sub) < 2 {
+			return match
+		}
+		content := reMrkdwnBoldStar.ReplaceAllString(sub[1], "$1")
+		content = reMrkdwnBoldUnder.ReplaceAllString(content, "$1")
+		return "*" + content + "*"
+	})
+
+	// 4. Bold: **text** and __text__ → *text*.
+	//    Process **bold** before __bold__ so nested cases like **__text__** work.
+	text = reMrkdwnBoldStar.ReplaceAllString(text, "*$1*")
+	text = reMrkdwnBoldUnder.ReplaceAllString(text, "*$1*")
+
+	// 5. Strikethrough: ~~text~~ → ~text~.
+	text = reMrkdwnStrike.ReplaceAllString(text, "~$1~")
+
+	// 6. Images before links (images have the ! prefix).
+	text = reMrkdwnImage.ReplaceAllString(text, "<$2|$1>")
+
+	// 7. Links: [text](url) → <url|text>.
+	text = reMrkdwnLink.ReplaceAllString(text, "<$2|$1>")
+
+	// 8. Horizontal rules → remove.
+	text = reMrkdwnHR.ReplaceAllString(text, "")
+
+	// 9. Unordered list markers → bullet character.
+	text = reMrkdwnULItem.ReplaceAllStringFunc(text, func(match string) string {
+		sub := reMrkdwnULItem.FindStringSubmatch(match)
+		indent := ""
+		if len(sub) > 1 {
+			indent = sub[1]
+		}
+		return indent + "\u2022 "
+	})
+
+	// 10. Restore inline code, then code blocks.
+	for i, code := range inlineCodes {
+		text = strings.Replace(text, inlineCodePlaceholder(i), code, 1)
+	}
+	for i, block := range codeBlocks {
+		text = strings.Replace(text, codePlaceholder(i), block, 1)
+	}
+
+	// Clean up excessive blank lines left by removed elements.
+	for strings.Contains(text, "\n\n\n") {
+		text = strings.ReplaceAll(text, "\n\n\n", "\n\n")
+	}
+
+	return text
+}
+
+// convertTables detects Markdown tables (consecutive lines matching |...|) and
+// wraps them in fenced code blocks so Slack renders them as preformatted text.
+func convertTables(text string) string {
+	lines := strings.Split(text, "\n")
+	var result []string
+	inTable := false
+
+	for i := 0; i < len(lines); i++ {
+		line := lines[i]
+
+		if reMrkdwnTableRow.MatchString(line) {
+			if !inTable {
+				inTable = true
+				result = append(result, "```")
+			}
+			// Skip separator rows (| --- | --- |) — they add no value in plain text.
+			if reMrkdwnTableSep.MatchString(line) {
+				continue
+			}
+			// Strip leading/trailing pipes and clean up cell content.
+			result = append(result, formatTableRow(line))
+		} else {
+			if inTable {
+				inTable = false
+				result = append(result, "```")
+			}
+			result = append(result, line)
+		}
+	}
+
+	// Close any trailing table.
+	if inTable {
+		result = append(result, "```")
+	}
+
+	return strings.Join(result, "\n")
+}
+
+// formatTableRow strips the outer pipes from a table row and trims each cell.
+func formatTableRow(line string) string {
+	// Remove leading and trailing pipe.
+	line = strings.TrimSpace(line)
+	line = strings.TrimPrefix(line, "|")
+	line = strings.TrimSuffix(line, "|")
+
+	cells := strings.Split(line, "|")
+	for i, cell := range cells {
+		cells[i] = strings.TrimSpace(cell)
+	}
+	return strings.Join(cells, " | ")
+}
+
+func codePlaceholder(idx int) string {
+	return fmt.Sprintf("\x00CODEBLOCK_%d\x00", idx)
+}
+
+func inlineCodePlaceholder(idx int) string {
+	return fmt.Sprintf("\x00INLINECODE_%d\x00", idx)
+}
diff --git a/pkg/channels/slack/parse_markdown_to_mrkdwn_test.go b/pkg/channels/slack/parse_markdown_to_mrkdwn_test.go
new file mode 100644
index 000000000..046bc10c3
--- /dev/null
+++ b/pkg/channels/slack/parse_markdown_to_mrkdwn_test.go
@@ -0,0 +1,247 @@
+package slack
+
+import "testing"
+
+func Test_markdownToSlackMrkdwn(t *testing.T) {
+	cases := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{
+			name:     "empty string",
+			input:    "",
+			expected: "",
+		},
+
+		// Bold
+		{
+			name:     "bold double asterisk",
+			input:    "this is **bold** text",
+			expected: "this is *bold* text",
+		},
+		{
+			name:     "bold double underscore",
+			input:    "this is __bold__ text",
+			expected: "this is *bold* text",
+		},
+		{
+			name:     "multiple bold segments",
+			input:    "**one** and **two**",
+			expected: "*one* and *two*",
+		},
+
+		// Italic (passthrough)
+		{
+			name:     "italic unchanged",
+			input:    "this is _italic_ text",
+			expected: "this is _italic_ text",
+		},
+
+		// Strikethrough
+		{
+			name:     "strikethrough",
+			input:    "this is ~~deleted~~ text",
+			expected: "this is ~deleted~ text",
+		},
+
+		// Headings
+		{
+			name:     "h1 heading",
+			input:    "# Main Title",
+			expected: "*Main Title*",
+		},
+		{
+			name:     "h3 heading",
+			input:    "### Sub Heading",
+			expected: "*Sub Heading*",
+		},
+		{
+			name:     "heading with inline bold collapses",
+			input:    "## **Bold Heading**",
+			expected: "*Bold Heading*",
+		},
+
+		// Links
+		{
+			name:     "markdown link",
+			input:    "visit [Google](https://google.com) today",
+			expected: "visit <https://google.com|Google> today",
+		},
+		{
+			name:     "link with special chars in text",
+			input:    "[click & go](https://example.com/path?a=1&b=2)",
+			expected: "<https://example.com/path?a=1&b=2|click & go>",
+		},
+
+		// Images
+		{
+			name:     "image to link",
+			input:    "![screenshot](https://example.com/img.png)",
+			expected: "<https://example.com/img.png|screenshot>",
+		},
+		{
+			name:     "image before link",
+			input:    "![img](https://a.com/1.png) and [link](https://b.com)",
+			expected: "<https://a.com/1.png|img> and <https://b.com|link>",
+		},
+
+		// Code blocks (passthrough)
+		{
+			name:     "fenced code block preserved",
+			input:    "```go\nfmt.Println(\"**not bold**\")\n```",
+			expected: "```go\nfmt.Println(\"**not bold**\")\n```",
+		},
+		{
+			name:     "inline code preserved",
+			input:    "use `**bold**` syntax",
+			expected: "use `**bold**` syntax",
+		},
+
+		// Blockquotes (passthrough)
+		{
+			name:     "blockquote unchanged",
+			input:    "> this is a quote",
+			expected: "> this is a quote",
+		},
+
+		// Horizontal rules
+		{
+			name:     "horizontal rule removed",
+			input:    "above\n---\nbelow",
+			expected: "above\n\nbelow",
+		},
+		{
+			name:     "asterisk hr removed",
+			input:    "above\n***\nbelow",
+			expected: "above\n\nbelow",
+		},
+
+		// Unordered lists
+		{
+			name:     "dash list to bullet",
+			input:    "- item one\n- item two",
+			expected: "\u2022 item one\n\u2022 item two",
+		},
+		{
+			name:     "asterisk list to bullet",
+			input:    "* item one\n* item two",
+			expected: "\u2022 item one\n\u2022 item two",
+		},
+		{
+			name:     "nested list indentation",
+			input:    "- top\n  - nested",
+			expected: "\u2022 top\n  \u2022 nested",
+		},
+
+		// Ordered lists (passthrough)
+		{
+			name:     "ordered list unchanged",
+			input:    "1. first\n2. second",
+			expected: "1. first\n2. second",
+		},
+
+		// Tables
+		{
+			name: "simple table",
+			input: "| Name | Age |\n| --- | --- |\n| Alice | 30 |\n| Bob | 25 |",
+			expected: "```\nName | Age\nAlice | 30\nBob | 25\n```",
+		},
+		{
+			name: "table with surrounding text",
+			input: "Here is a table:\n\n| A | B |\n| --- | --- |\n| 1 | 2 |\n\nEnd.",
+			expected: "Here is a table:\n\n```\nA | B\n1 | 2\n```\n\nEnd.",
+		},
+
+		// Mixed content
+		{
+			name: "realistic LLM output",
+			input: "## Summary\n\nHere are the **key points**:\n\n- First item with `code`\n- Second item with [a link](https://example.com)\n- ~~Removed~~ item\n\n### Details\n\n| Feature | Status |\n| --- | --- |\n| Auth | Done |\n| API | WIP |\n\n> Note: check the docs.",
+			expected: "*Summary*\n\nHere are the *key points*:\n\n\u2022 First item with `code`\n\u2022 Second item with <https://example.com|a link>\n\u2022 ~Removed~ item\n\n*Details*\n\n```\nFeature | Status\nAuth | Done\nAPI | WIP\n```\n\n> Note: check the docs.",
+		},
+
+		// Edge cases
+		{
+			name:     "single asterisk not converted",
+			input:    "this * is not * bold",
+			expected: "this * is not * bold",
+		},
+		{
+			name:     "bold inside code block not converted",
+			input:    "```\n**bold** inside code\n```",
+			expected: "```\n**bold** inside code\n```",
+		},
+		{
+			name:     "link inside inline code not converted",
+			input:    "use `[text](url)` for links",
+			expected: "use `[text](url)` for links",
+		},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			actual := markdownToSlackMrkdwn(tc.input)
+			if actual != tc.expected {
+				t.Errorf("\ninput:    %q\nexpected: %q\nactual:   %q", tc.input, tc.expected, actual)
+			}
+		})
+	}
+}
+
+func Test_convertTables(t *testing.T) {
+	cases := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{
+			name:     "no table",
+			input:    "just some text",
+			expected: "just some text",
+		},
+		{
+			name:     "simple two column",
+			input:    "| A | B |\n| --- | --- |\n| 1 | 2 |",
+			expected: "```\nA | B\n1 | 2\n```",
+		},
+		{
+			name:     "three columns with alignment markers",
+			input:    "| Left | Center | Right |\n| :--- | :---: | ---: |\n| a | b | c |",
+			expected: "```\nLeft | Center | Right\na | b | c\n```",
+		},
+		{
+			name:     "multiple tables separated by text",
+			input:    "| A | B |\n| - | - |\n| 1 | 2 |\n\ntext\n\n| C | D |\n| - | - |\n| 3 | 4 |",
+			expected: "```\nA | B\n1 | 2\n```\n\ntext\n\n```\nC | D\n3 | 4\n```",
+		},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			actual := convertTables(tc.input)
+			if actual != tc.expected {
+				t.Errorf("\ninput:    %q\nexpected: %q\nactual:   %q", tc.input, tc.expected, actual)
+			}
+		})
+	}
+}
+
+func Test_formatTableRow(t *testing.T) {
+	cases := []struct {
+		input    string
+		expected string
+	}{
+		{"| A | B |", "A | B"},
+		{"| one | two | three |", "one | two | three"},
+		{"|  spaced  |  cells  |", "spaced | cells"},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.input, func(t *testing.T) {
+			actual := formatTableRow(tc.input)
+			if actual != tc.expected {
+				t.Errorf("formatTableRow(%q) = %q, want %q", tc.input, actual, tc.expected)
+			}
+		})
+	}
+}
diff --git a/pkg/channels/slack/slack.go b/pkg/channels/slack/slack.go
index acd857a06..99a50b5fc 100644
--- a/pkg/channels/slack/slack.go
+++ b/pkg/channels/slack/slack.go
@@ -119,7 +119,7 @@ func (c *SlackChannel) Send(ctx context.Context, msg bus.OutboundMessage) error
 	}
 
 	opts := []slack.MsgOption{
-		slack.MsgOptionText(msg.Content, false),
+		slack.MsgOptionText(markdownToSlackMrkdwn(msg.Content), false),
 	}
 
 	if msg.ReplyToMessageID != "" && threadTS == "" {