diff --git a/src/agentops/cli/report_commands.py b/src/agentops/cli/report_commands.py
index 93c4ac3c..81b18212 100644
--- a/src/agentops/cli/report_commands.py
+++ b/src/agentops/cli/report_commands.py
@@ -7,7 +7,6 @@
 
 import typer
 
-from agentops.cli._planned import _planned_command
 from agentops.utils.logging import get_logger
 
 log = get_logger(__name__)
@@ -72,12 +71,129 @@ def cmd_report(
 
 
 @report_app.command("show")
-def cmd_report_show() -> None:
-    """View reports in table format (planned)."""
-    _planned_command("agentops report show")
+def cmd_report_show(
+    results_in: Annotated[
+        Path | None,
+        typer.Option(
+            "--in",
+            help=(
+                "Path to results.json. "
+                "If omitted, uses .agentops/results/latest/results.json"
+            ),
+        ),
+    ] = None,
+    report_format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Report format to display: md or html."),
+    ] = "md",
+) -> None:
+    """Display an existing evaluation report."""
+    import webbrowser
+
+    resolved_in = (results_in or DEFAULT_REPORT_INPUT).resolve()
+    results_dir = resolved_in.parent
+
+    if report_format == "html":
+        html_path = results_dir / "report.html"
+        if not html_path.exists():
+            typer.echo(
+                f"Error: report.html not found in {results_dir}. "
+                "Regenerate with: agentops report --format html",
+                err=True,
+            )
+            raise typer.Exit(code=1)
+        typer.echo(f"Opening {html_path} in browser...")
+        webbrowser.open(html_path.as_uri())
+    else:
+        md_path = results_dir / "report.md"
+        if not md_path.exists():
+            typer.echo(
+                f"Error: report.md not found in {results_dir}. "
+                "Regenerate with: agentops report",
+                err=True,
+            )
+            raise typer.Exit(code=1)
+        typer.echo(md_path.read_text(encoding="utf-8"))
 
 
 @report_app.command("export")
-def cmd_report_export() -> None:
-    """Export reports in JSON/Markdown/CSV formats (planned)."""
-    _planned_command("agentops report export")
+def cmd_report_export(
+    results_in: Annotated[
+        Path | None,
+        typer.Option(
+            "--in",
+            help=(
+                "Path to results.json. "
+                "If omitted, uses .agentops/results/latest/results.json"
+            ),
+        ),
+    ] = None,
+    output: Annotated[
+        Path | None,
+        typer.Option("--out", "-o", help="Output file path."),
+    ] = None,
+    export_format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Export format: json, csv, or md."),
+    ] = "json",
+) -> None:
+    """Export evaluation results to JSON, CSV, or Markdown."""
+    import csv
+    import io
+    import json
+
+    resolved_in = (results_in or DEFAULT_REPORT_INPUT).resolve()
+
+    if not resolved_in.exists():
+        typer.echo(f"Error: results.json not found: {resolved_in}", err=True)
+        raise typer.Exit(code=1)
+
+    data = json.loads(resolved_in.read_text(encoding="utf-8"))
+
+    if export_format == "json":
+        content = json.dumps(data, indent=2)
+    elif export_format == "csv":
+        metrics = data.get("metrics", [])
+        buf = io.StringIO()
+        writer = csv.writer(buf)
+        writer.writerow(["metric", "value"])
+        for m in metrics:
+            writer.writerow([m.get("name", ""), m.get("value", "")])
+
+        # Add row-level metrics if available
+        row_metrics = data.get("row_metrics", [])
+        if row_metrics:
+            writer.writerow([])
+            # Collect all metric names across rows
+            metric_names: list[str] = []
+            for row in row_metrics:
+                for m in row.get("metrics", []):
+                    if m["name"] not in metric_names:
+                        metric_names.append(m["name"])
+            writer.writerow(["row_index"] + metric_names)
+            for row in row_metrics:
+                values = {m["name"]: m["value"] for m in row.get("metrics", [])}
+                writer.writerow(
+                    [row.get("row_index", "")]
+                    + [values.get(n, "") for n in metric_names]
+                )
+        content = buf.getvalue()
+    elif export_format == "md":
+        md_path = resolved_in.parent / "report.md"
+        if not md_path.exists():
+            typer.echo(
+                "Error: report.md not found. Regenerate with: agentops report",
+                err=True,
+            )
+            raise typer.Exit(code=1)
+        content = md_path.read_text(encoding="utf-8")
+    else:
+        typer.echo("Error: --format must be json, csv, or md.", err=True)
+        raise typer.Exit(code=1)
+
+    if output:
+        output.parent.mkdir(parents=True, exist_ok=True)
+        output.write_text(content, encoding="utf-8")
+        typer.echo(f"Exported to {output}")
+    else:
+        typer.echo(content)
diff --git a/tests/unit/test_report_commands.py b/tests/unit/test_report_commands.py
new file mode 100644
index 00000000..140322d0
--- /dev/null
+++ b/tests/unit/test_report_commands.py
@@ -0,0 +1,250 @@
+"""Tests for report show and report export commands."""
+
+from __future__ import annotations
+
+import csv
+import io
+import json
+from pathlib import Path
+
+from typer.testing import CliRunner
+
+from agentops.cli.app import app
+
+runner = CliRunner()
+
+
+def _create_results_dir(tmp_path: Path, *, with_html: bool = False) -> Path:
+    """Create a results directory with results.json and report.md."""
+    results_dir = tmp_path / ".agentops" / "results" / "latest"
+    results_dir.mkdir(parents=True)
+
+    results = {
+        "version": 1,
+        "status": "completed",
+        "bundle": {"name": "test_bundle", "path": "bundles/test.yaml"},
+        "dataset": {"name": "test_dataset", "path": "datasets/test.yaml"},
+        "execution": {
+            "backend": "foundry",
+            "command": "test",
+            "started_at": "2026-04-07T10:00:00Z",
+            "finished_at": "2026-04-07T10:01:00Z",
+            "duration_seconds": 60.0,
+            "exit_code": 0,
+        },
+        "metrics": [
+            {"name": "CoherenceEvaluator", "value": 4.5},
+            {"name": "RelevanceEvaluator", "value": 5.0},
+            {"name": "samples_evaluated", "value": 3.0},
+        ],
+        "row_metrics": [
+            {
+                "row_index": 1,
+                "metrics": [
+                    {"name": "CoherenceEvaluator", "value": 4.0},
+                    {"name": "RelevanceEvaluator", "value": 5.0},
+                ],
+            },
+            {
+                "row_index": 2,
+                "metrics": [
+                    {"name": "CoherenceEvaluator", "value": 5.0},
+                    {"name": "RelevanceEvaluator", "value": 5.0},
+                ],
+            },
+        ],
+        "item_evaluations": [],
+        "thresholds": [],
+        "summary": {
+            "metrics_count": 3,
+            "thresholds_count": 0,
+            "thresholds_passed": 0,
+            "thresholds_failed": 0,
+            "overall_passed": True,
+        },
+    }
+    (results_dir / "results.json").write_text(
+        json.dumps(results, indent=2), encoding="utf-8"
+    )
+    (results_dir / "report.md").write_text(
+        "# Test Report\n\nAll passed.\n", encoding="utf-8"
+    )
+    if with_html:
+        (results_dir / "report.html").write_text(
+            "<html><body>Report</body></html>", encoding="utf-8"
+        )
+    return results_dir
+
+
+# ---------------------------------------------------------------------------
+# report show
+# ---------------------------------------------------------------------------
+
+
+class TestReportShow:
+    def test_show_md(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            ["report", "show", "--in", str(results_dir / "results.json")],
+        )
+        assert result.exit_code == 0
+        assert "# Test Report" in result.stdout
+        assert "All passed." in result.stdout
+
+    def test_show_md_default_format(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            ["report", "show", "--in", str(results_dir / "results.json")],
+        )
+        assert result.exit_code == 0
+        assert "# Test Report" in result.stdout
+
+    def test_show_md_missing(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        (results_dir / "report.md").unlink()
+        result = runner.invoke(
+            app,
+            ["report", "show", "--in", str(results_dir / "results.json")],
+        )
+        assert result.exit_code == 1
+
+    def test_show_html_missing(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "show",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "html",
+            ],
+        )
+        assert result.exit_code == 1
+        assert "report.html not found" in (result.stdout + result.stderr)
+
+
+# ---------------------------------------------------------------------------
+# report export
+# ---------------------------------------------------------------------------
+
+
+class TestReportExport:
+    def test_export_json_stdout(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "json",
+            ],
+        )
+        assert result.exit_code == 0
+        data = json.loads(result.stdout)
+        assert data["status"] == "completed"
+        assert len(data["metrics"]) == 3
+
+    def test_export_json_to_file(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        out_file = tmp_path / "export.json"
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--out",
+                str(out_file),
+            ],
+        )
+        assert result.exit_code == 0
+        assert out_file.exists()
+        data = json.loads(out_file.read_text(encoding="utf-8"))
+        assert data["status"] == "completed"
+
+    def test_export_csv_stdout(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "csv",
+            ],
+        )
+        assert result.exit_code == 0
+        reader = csv.reader(io.StringIO(result.stdout))
+        rows = list(reader)
+        # Header + 3 metrics + blank + row header + 2 row metrics
+        assert rows[0] == ["metric", "value"]
+        assert rows[1][0] == "CoherenceEvaluator"
+
+    def test_export_csv_to_file(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        out_file = tmp_path / "export.csv"
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "csv",
+                "--out",
+                str(out_file),
+            ],
+        )
+        assert result.exit_code == 0
+        assert out_file.exists()
+        content = out_file.read_text(encoding="utf-8")
+        assert "CoherenceEvaluator" in content
+
+    def test_export_md_stdout(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "md",
+            ],
+        )
+        assert result.exit_code == 0
+        assert "# Test Report" in result.stdout
+
+    def test_export_invalid_format(self, tmp_path: Path) -> None:
+        results_dir = _create_results_dir(tmp_path)
+        result = runner.invoke(
+            app,
+            [
+                "report",
+                "export",
+                "--in",
+                str(results_dir / "results.json"),
+                "--format",
+                "xml",
+            ],
+        )
+        assert result.exit_code == 1
+
+    def test_export_missing_results(self, tmp_path: Path) -> None:
+        result = runner.invoke(
+            app,
+            ["report", "export", "--in", str(tmp_path / "nonexistent.json")],
+        )
+        assert result.exit_code == 1