diff --git a/src/investo/cli.py b/src/investo/cli.py
index cf1404c..43584fc 100644
--- a/src/investo/cli.py
+++ b/src/investo/cli.py
@@ -316,17 +316,22 @@ def _cmd_analyze(args: argparse.Namespace) -> int:
         print(json.dumps(report.model_dump(), indent=2, default=str))
 
     if want_html:
-        from .export import save_html
+        from .export import file_url, save_html
         out = save_html(report, args.html)  # args.html is None when the flag is bare
         print(f"Wrote HTML report to {out}")
+        print(f"  Open: {file_url(out)}")  # clickable — opens in the browser on click
 
     if want_pdf:
-        from .export import PdfExportError, save_pdf
+        from .export import PdfExportError, file_url, save_pdf
         try:
             out, engine, warnings = save_pdf(report, args.pdf)
             for w in warnings:
                 print(f"warning: {w}", file=sys.stderr)
             print(f"Wrote PDF report to {out}  ({engine})")
+            print(f"  Open: {file_url(out)}")  # the pdf
+            sidecar = out.with_suffix(".html")
+            if sidecar.exists():
+                print(f"  HTML: {file_url(sidecar)}")  # the html written alongside it
         except PdfExportError as exc:
             # The .html sidecar is still on disk; a failed export is still a failed command.
             print(f"error: PDF export failed.\n{exc}", file=sys.stderr)
@@ -336,9 +341,10 @@ def _cmd_analyze(args: argparse.Namespace) -> int:
     # (--html / --pdf already produce a document) or opted out with --no-html. The announcement
     # goes to stderr so `investo analyze X --json` stays pipeable on stdout.
     if not (want_html or want_pdf or getattr(args, "no_html", False)):
-        from .export import save_html
+        from .export import file_url, save_html
         out = save_html(report, None)
         print(f"Wrote HTML report to {out}", file=sys.stderr)
+        print(f"  Open: {file_url(out)}", file=sys.stderr)
 
     if not (args.json or want_html or want_pdf):
         print(render_report(report))
diff --git a/src/investo/export.py b/src/investo/export.py
index cc6a437..3a4927f 100644
--- a/src/investo/export.py
+++ b/src/investo/export.py
@@ -32,6 +32,17 @@
 _UNSAFE = str.maketrans(dict.fromkeys('<>:"/\\|?*&%', "-"))
 
 
+def file_url(path: str | os.PathLike) -> str:
+    """A clickable, absolute ``file://`` URL for a local path, percent-escaped.
+
+    ``resolve().as_uri()`` is the same construction the headless-Chrome path uses to load the
+    report, so clicking this link opens exactly the file that was written (a space becomes ``%20``,
+    Windows drive letters get ``file:///C:/…``). Callers surface it so a reader can open the report
+    without hunting for the path.
+    """
+    return Path(path).resolve().as_uri()
+
+
 class PdfExportError(RuntimeError):
     """Raised when no PDF backend could produce a file. The message names the remedies."""
 
@@ -139,9 +150,9 @@ def _chrome_pdf(browser: Path, html: str, out_path: Path, timeout: float) -> str
         src.write_text(html, encoding="utf-8")
         profile = tmp_dir / "profile"  # (3) throwaway profile, see below
 
-        # (2) resolve().as_uri() builds file:///C:/... and percent-escapes spaces. A hand-built
-        #     "file://" + str(path) gets both the slashes and the spaces wrong on Windows.
-        url = src.resolve().as_uri()
+        # (2) file_url() -> resolve().as_uri() builds file:///C:/... and percent-escapes spaces. A
+        #     hand-built "file://" + str(path) gets both the slashes and the spaces wrong on Windows.
+        url = file_url(src)
 
         cmd = [
             str(browser),
diff --git a/src/investo/models.py b/src/investo/models.py
index d68dcf5..94f42b5 100644
--- a/src/investo/models.py
+++ b/src/investo/models.py
@@ -548,12 +548,25 @@ class AiSignals(_Base):
 # --------------------------------------------------------------------------------------
 # Analysis tools: export, technicals, DCF sensitivity, multi-company compare
 # --------------------------------------------------------------------------------------
+class ExportedFile(_Base):
+    """One written artifact and how to open it."""
+
+    path: str
+    file_url: str  # clickable file:// URL — opens the file in the default app on click
+    format: Literal["pdf", "html"]
+    bytes: int = 0
+
+
 class ExportResult(_Base):
     path: str
+    file_url: str | None = None  # clickable file:// URL of `path`
     format: Literal["pdf", "html"]
     bytes: int = 0
     engine: str | None = None  # e.g. "chrome.exe (headless)" | "playwright-chromium"
     warnings: list[str] = Field(default_factory=list)
+    # Every artifact this call wrote, primary first — a PDF export also lists its HTML sidecar,
+    # so a caller has a clickable location for the html and the pdf.
+    files: list[ExportedFile] = Field(default_factory=list)
 
 
 class TechnicalSnapshot(_Base):
@@ -683,6 +696,7 @@ class AnalysisReport(_Base):
     # Auto-export metadata: set when analyze_company writes an HTML report so a client can open
     # it without a second tool call. Optional — the report is complete without them.
     html_report_path: str | None = None
+    html_report_url: str | None = None   # clickable file:// URL of html_report_path
     generated_at: str | None = None      # ISO-8601 UTC timestamp of when the report was written
     investo_version: str | None = None
     html_bytes: int | None = None
diff --git a/src/investo/server.py b/src/investo/server.py
index 8c19d3c..b64b0f9 100644
--- a/src/investo/server.py
+++ b/src/investo/server.py
@@ -28,6 +28,7 @@
     CompanyProfile,
     DCFResult,
     DcfSensitivity,
+    ExportedFile,
     ExportResult,
     Financials,
     FundamentalTrend,
@@ -136,13 +137,14 @@ def _attach_html_report(report: AnalysisReport) -> None:
     from datetime import datetime, timezone
 
     from . import __version__
-    from .export import default_filename, save_html
+    from .export import default_filename, file_url, save_html
 
     report.generated_at = datetime.now(timezone.utc).isoformat(timespec="seconds")
     report.investo_version = __version__
     try:
         out = save_html(report, _safe_export_path(default_filename(report, "html"), "html"))
         report.html_report_path = str(out)
+        report.html_report_url = file_url(out)  # clickable — opens the report on click
         report.html_bytes = out.stat().st_size
     except Exception as exc:  # noqa: BLE001 — the convenience export must never sink the analysis
         _log.warning("automatic HTML export failed: %s", exc)
@@ -411,25 +413,31 @@ def export_report(
     a caller cannot write outside it.
     """
     from .analysis.report import analyze
-    from .export import PdfExportError, html_to_pdf, save_html
+    from .export import PdfExportError, file_url, html_to_pdf, save_html
+
+    def _exported(p, fmt: Literal["pdf", "html"]) -> ExportedFile:
+        return ExportedFile(path=str(p), file_url=file_url(p), format=fmt, bytes=p.stat().st_size)
 
     out = _safe_export_path(path, format)
     report = analyze(query, market)
 
     if format == "html":
-        written = save_html(report, out)
-        return ExportResult(path=str(written), format="html",
-                            bytes=written.stat().st_size, engine="renderer")
+        f = _exported(save_html(report, out), "html")
+        return ExportResult(path=f.path, file_url=f.file_url, format="html",
+                            bytes=f.bytes, engine="renderer", files=[f])
 
     from .render import render_html
     html = render_html(report)
-    out.with_suffix(".html").write_text(html, encoding="utf-8")  # sidecar survives a PDF failure
+    sidecar = out.with_suffix(".html")
+    sidecar.write_text(html, encoding="utf-8")  # sidecar survives a PDF failure
     try:
         engine, warnings = html_to_pdf(html, out)
     except PdfExportError as exc:
         raise ValueError(f"{exc}") from exc
-    return ExportResult(path=str(out), format="pdf", bytes=out.stat().st_size,
-                        engine=engine, warnings=warnings)
+    # Return a clickable location for the pdf and its html sidecar (primary first).
+    pdf, html_side = _exported(out, "pdf"), _exported(sidecar, "html")
+    return ExportResult(path=pdf.path, file_url=pdf.file_url, format="pdf", bytes=pdf.bytes,
+                        engine=engine, warnings=warnings, files=[pdf, html_side])
 
 
 @mcp.tool(title="Full analysis", annotations=_WRITE)
diff --git a/tests/test_cli.py b/tests/test_cli.py
index 3875b99..e3bb4e7 100644
--- a/tests/test_cli.py
+++ b/tests/test_cli.py
@@ -40,6 +40,7 @@ def test_no_output_flag_prints_the_terminal_report_and_auto_writes_html(tmp_path
     written = list(tmp_path.glob("investo-KPITTECH.NS-*.html"))
     assert len(written) == 1
     assert "Wrote HTML report" in captured.err
+    assert "file://" in captured.err  # clickable location, announced on stderr
 
 
 def test_json_flag_emits_valid_json_on_a_clean_stdout(tmp_path, monkeypatch, capsys):
@@ -72,7 +73,9 @@ def test_html_flag_writes_a_file_and_creates_parents(tmp_path, capsys):
     assert _run(["analyze", "KPIT", "--html", str(target)]) == 0
     assert target.exists()
     assert target.read_text(encoding="utf-8").startswith("<!doctype html>")
-    assert "Wrote HTML report" in capsys.readouterr().out
+    out = capsys.readouterr().out
+    assert "Wrote HTML report" in out
+    assert "file://" in out and "report.html" in out  # a clickable location to open it
 
 
 def test_bare_html_flag_uses_a_default_name(tmp_path, monkeypatch, capsys):
@@ -95,7 +98,9 @@ def test_pdf_success_reports_the_engine(tmp_path, monkeypatch, capsys):
     monkeypatch.setattr("investo.export.save_pdf",
                         lambda report, path: (tmp_path / "k.pdf", "chrome (headless)", []))
     assert _run(["analyze", "KPIT", "--pdf", str(tmp_path / "k.pdf")]) == 0
-    assert "chrome (headless)" in capsys.readouterr().out
+    out = capsys.readouterr().out
+    assert "chrome (headless)" in out
+    assert "file://" in out and "k.pdf" in out  # clickable location for the pdf
 
 
 def test_pdf_failure_exits_two_and_keeps_the_html(tmp_path, monkeypatch, capsys):
diff --git a/tests/test_export.py b/tests/test_export.py
index 324aba5..2b12f95 100644
--- a/tests/test_export.py
+++ b/tests/test_export.py
@@ -73,6 +73,17 @@ def test_windows_style_path_becomes_a_percent_escaped_file_uri():
     assert "%20" in uri  # the space is escaped, not left raw
 
 
+def test_file_url_is_an_absolute_clickable_uri(tmp_path):
+    # The clickable location returned to the user: absolute, percent-escaped, opens on click.
+    target = tmp_path / "a b" / "investo-report.html"
+    target.parent.mkdir(parents=True)
+    target.write_text("<html></html>", encoding="utf-8")
+    url = export.file_url(target)
+    assert url.startswith("file:///")
+    assert "%20" in url  # the space in the directory name is escaped
+    assert url.endswith("investo-report.html")
+
+
 # --------------------------------------------------------------------------------------
 # Headless Chrome path — stubbed subprocess, never a real launch
 # --------------------------------------------------------------------------------------
diff --git a/tests/test_server_tools.py b/tests/test_server_tools.py
index 47b5d9e..b600427 100644
--- a/tests/test_server_tools.py
+++ b/tests/test_server_tools.py
@@ -110,6 +110,9 @@ def test_attach_html_report_writes_into_the_sandbox_and_records_metadata(tmp_pat
     assert written.read_text(encoding="utf-8").startswith("<!doctype html>")
     assert report.html_bytes and report.html_bytes > 0
     assert report.investo_version and report.generated_at
+    # A clickable location so a client can open the report without a second call.
+    assert report.html_report_url and report.html_report_url.startswith("file:///")
+    assert report.html_report_url == written.resolve().as_uri()
 
 
 # --------------------------------------------------------------------------------------