fix(tests): strip ANSI escapes from --help output before substring asserts

Typer/rich emits ANSI styling in CI's --help output (e.g. `--metrics`
becomes `-\x1b[0m\x1b[1;36m-metrics`), so literal substring asserts
like `assert "--metrics" in result.output` fail. Locally the test runner
auto-detects no-TTY and produces plain text, masking the issue.

Add a small `_clean()` helper per test file that strips ANSI escape
codes (`\x1b\[[0-9;]*m`) before substring containment checks.

tests/test_cli_admin_metrics.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.admin import admin_app
 
 
@@ -9,6 +14,6 @@ def test_admin_metrics_subcommands_present():
     runner = CliRunner()
     result = runner.invoke(admin_app, ["metrics", "--help"])
     assert result.exit_code == 0
-    assert "import" in result.output
-    assert "export" in result.output
-    assert "validate" in result.output
+    assert "import" in _clean(result.output)
+    assert "export" in _clean(result.output)
+    assert "validate" in _clean(result.output)

tests/test_cli_catalog_metrics.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.catalog import catalog_app
 
 
@@ -9,8 +14,8 @@ def test_catalog_metrics_help():
     runner = CliRunner()
     result = runner.invoke(catalog_app, ["--help"])
     assert result.exit_code == 0
-    assert "--metrics" in result.output
-    assert "--show" in result.output
+    assert "--metrics" in _clean(result.output)
+    assert "--show" in _clean(result.output)
 
 
 def test_catalog_default_still_works():
@@ -20,4 +25,4 @@ def test_catalog_default_still_works():
     result = runner.invoke(catalog_app, ["--help"])
     assert result.exit_code == 0
     # No traceback
-    assert "Traceback" not in result.output
+    assert "Traceback" not in _clean(result.output)

tests/test_cli_diagnose_system.py

@@ -3,6 +3,11 @@
 from typer.testing import CliRunner
 from cli.commands.diagnose import diagnose_app
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 runner = CliRunner()
 
 
@@ -15,7 +20,7 @@ def test_diagnose_help_lists_system():
     """Top-level diagnose help should mention the `system` subcommand."""
     result = runner.invoke(diagnose_app, ["--help"])
     assert result.exit_code == 0
-    assert "system" in result.output
+    assert "system" in _clean(result.output)
 
 
 def test_diagnose_default_still_works():
@@ -24,4 +29,4 @@ def test_diagnose_default_still_works():
     result = runner.invoke(diagnose_app, [])
     # Either runs successfully or fails for unrelated reasons (no server etc).
     # We just want to verify no traceback from the addition.
-    assert "Traceback" not in (result.output + (result.stderr or ""))
+    assert "Traceback" not in (_clean(result.output) + _clean(result.stderr or ''))

tests/test_cli_init.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.init import init_app
 
 runner = CliRunner()
@@ -42,10 +47,10 @@ def _make_api_get():
 def test_init_help():
     result = runner.invoke(init_app, ["--help"])
     assert result.exit_code == 0
-    assert "--server-url" in result.output
-    assert "--token" in result.output
-    assert "--force" in result.output
-    assert "--workspace" in result.output
+    assert "--server-url" in _clean(result.output)
+    assert "--token" in _clean(result.output)
+    assert "--force" in _clean(result.output)
+    assert "--workspace" in _clean(result.output)
 
 
 def test_init_writes_expected_files(tmp_path, monkeypatch):
@@ -130,7 +135,7 @@ def test_init_partial_state_friendly_exit(tmp_path, monkeypatch):
         "--workspace", str(workspace),
     ])
     assert result.exit_code == 1
-    assert "Traceback" not in (result.output + (result.stderr or ""))
+    assert "Traceback" not in (_clean(result.output) + _clean(result.stderr or ''))
 
 
 def test_init_auth_failed_on_401(tmp_path, monkeypatch):

tests/test_cli_pull.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.pull import pull_app
 
 runner = CliRunner()
@@ -10,9 +15,9 @@ runner = CliRunner()
 def test_pull_help():
     result = runner.invoke(pull_app, ["--help"])
     assert result.exit_code == 0
-    assert "--quiet" in result.output
-    assert "--json" in result.output
-    assert "--dry-run" in result.output
+    assert "--quiet" in _clean(result.output)
+    assert "--json" in _clean(result.output)
+    assert "--dry-run" in _clean(result.output)
 
 
 def test_pull_no_server_friendly_exit(tmp_path, monkeypatch):
@@ -23,4 +28,4 @@ def test_pull_no_server_friendly_exit(tmp_path, monkeypatch):
     result = runner.invoke(pull_app, [])
     # Either exit 1 with hint, or exit 0 if a default server URL applies.
     # Either way, there must be no Python traceback in stderr/stdout.
-    assert "Traceback" not in (result.output + (result.stderr or ""))
+    assert "Traceback" not in (_clean(result.output) + _clean(result.stderr or ''))

tests/test_cli_push.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.push import push_app
 
 runner = CliRunner()
@@ -10,9 +15,9 @@ runner = CliRunner()
 def test_push_help():
     result = runner.invoke(push_app, ["--help"])
     assert result.exit_code == 0
-    assert "--quiet" in result.output
-    assert "--json" in result.output
-    assert "--dry-run" in result.output
+    assert "--quiet" in _clean(result.output)
+    assert "--json" in _clean(result.output)
+    assert "--dry-run" in _clean(result.output)
 
 
 def test_push_no_sessions_no_mkdir(tmp_path, monkeypatch):

tests/test_cli_snapshot_create.py

@@ -2,6 +2,11 @@
 
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.snapshot import snapshot_app
 
 
@@ -19,7 +24,7 @@ def test_snapshot_create_help():
         "--no-estimate",
         "--force",
     ]:
-        assert flag in result.output
+        assert flag in _clean(result.output)
 
 
 def test_snapshot_create_no_duckdb_friendly_exit(tmp_path, monkeypatch):

tests/test_cli_status.py

@@ -3,6 +3,11 @@
 import json
 from typer.testing import CliRunner
 
+# CI-safety: Typer/rich emits ANSI escapes in --help output. Strip before asserts.
+_ANSI_RE = __import__("re").compile(r"\x1b\[[0-9;]*m")
+def _clean(s: str) -> str:
+    return _ANSI_RE.sub("", s)
+
 from cli.commands.status import status_app
 
 runner = CliRunner()
@@ -15,7 +20,7 @@ def test_status_uninitialized_workspace(tmp_path, monkeypatch):
     assert result.exit_code in (0, 1)
     out = result.output.lower()
     assert "no" in out  # "Initialized: no" or similar
-    assert "agnes init" in result.output  # hint to initialize
+    assert "agnes init" in _clean(result.output)  # hint to initialize
 
 
 def test_status_initialized_workspace(tmp_path, monkeypatch):
@@ -31,7 +36,7 @@ def test_status_initialized_workspace(tmp_path, monkeypatch):
     assert result.exit_code == 0
     out = result.output.lower()
     assert "yes" in out  # "Initialized: yes"
-    assert "1" in result.output  # one parquet
+    assert "1" in _clean(result.output)  # one parquet
 
 
 def test_status_json(tmp_path, monkeypatch):