failproofai 1.0.12-beta.2 → 1.0.12

package/.next/standalone/fp-cloud-cli/tests/test_output.py

@@ -51,7 +51,7 @@ def test_print_table_writes_to_stdout(capsys):
     output.configure(no_color=True, quiet=False)
     output.print_table(["a", "b"], [["1", "2"]], title="Demo")
     out = capsys.readouterr().out
-    assert "Demo" in out
+    assert "DEMO" in out
     assert "1" in out and "2" in out
 
 
@@ -91,7 +91,7 @@ def test_render_events_box_to_stdout(capsys):
     output.render_events([_event("2026-06-22T12:21:54Z")], order="desc")
     captured = capsys.readouterr()
     # The box (human data view) goes to STDOUT; the title carries count, direction and date.
-    assert "events" in captured.out
+    assert "EVENTS" in captured.out
     assert "newest first" in captured.out
     assert "2026-06-22" in captured.out
     assert captured.err == ""
@@ -182,7 +182,7 @@ def test_status_color_enum():
 
 
 def test_short_session():
-    assert output._short_session("sess-20260615-fcf97e01") == "sess-…fcf97e01"
+    assert output._short_session("sess-20260615-fcf97e01") == "sess-20260615-fcf97e01"
     assert output._short_session("sess-20260615-fcf97e01", full=True) == "sess-20260615-fcf97e01"
     assert output._short_session("nonnum-3-b9d9de") == "nonnum-3-b9d9de"  # short → intact
     assert output._short_session("short") == "short"
@@ -208,9 +208,9 @@ def test_render_sessions_box_has_no_scores(capsys):
     _wide_stdout()
     output.render_sessions([_sample_eval()])
     out = capsys.readouterr().out
-    assert "sessions" in out and "newest first" in out and "2026-06-22" in out
+    assert "SESSIONS" in out and "newest first" in out and "2026-06-22" in out
     assert "agent-codegen" in out and "done" in out
-    assert "12:56:31" in out and "sess-…fcf97e01" in out
+    assert "12:56:31" in out and "sess-20260615-fcf97e01" in out
     assert "coherence" not in out and "scores" not in out  # scores moved to evals
     output.configure(no_color=True, quiet=False)
 
@@ -219,8 +219,8 @@ def test_render_evals_box_has_scores(capsys):
     _wide_stdout()
     output.render_evals([_sample_eval()])
     out = capsys.readouterr().out
-    assert "evals" in out and "scores" in out and "coherence" in out
-    assert "agent-codegen" in out and "sess-…fcf97e01" in out
+    assert "EVALS" in out and "scores" in out and "coherence" in out
+    assert "agent-codegen" in out and "sess-20260615-fcf97e01" in out
     output.configure(no_color=True, quiet=False)
 
 
@@ -296,7 +296,7 @@ def test_render_sessions_expanded_lists_full_roster(capsys):
     assert "agent-testgen" in out and "9 ev" in out
     assert "●" not in out                                  # uniform: no special "root" marker
     assert "├" in out and "└" in out                       # uniform tree glyphs, └ closes the list
-    assert "sessions · 1" in out                           # panel count = sessions, not rows
+    assert "SESSIONS · 1" in out                           # count = sessions, not rows
     output.configure(no_color=True, quiet=False)
 
 
@@ -403,8 +403,8 @@ def test_render_errors_box(capsys):
     _wide_stdout()
     output.render_errors([_errevent("error", "RateLimitError: upstream timed out")])
     out = capsys.readouterr().out
-    assert "errors" in out and "newest first" in out
-    assert "agent-orderbot" in out and "sess-…4b90b240" in out  # truncated session (shared last-8)
+    assert "ERRORS" in out and "newest first" in out
+    assert "agent-orderbot" in out and "sess-20260622-4b90b240" in out
     assert "RateLimitError: upstream timed out" in out          # server-computed summary field
     assert "17:51:34" in out
     output.configure(no_color=True, quiet=False)
@@ -479,7 +479,7 @@ def test_render_org_perms(capsys):
     out = capsys.readouterr().out
     # header leads with the org name + slug + role; the count moved into the box title
     assert "Globex Corp" in out and "globex" in out and "role admin" in out
-    assert "permissions · 5 · Globex Corp" in out
+    assert "PERMISSIONS · 5 · Globex Corp" in out
     assert "dashboards" in out and "keys" in out and "agent" in out
     output.configure(no_color=True, quiet=False)
 
@@ -549,7 +549,7 @@ def test_render_value_list_narrow_caps_columns(capsys):
 
 
 def test_short_id():
-    assert output._short_id("1f58376d-7947-9826") == "1f58…9826"
+    assert output._short_id("1f58376d-7947-9826") == "1f58376d-7947-9826"
     assert output._short_id("short") == "short"
 
 
@@ -585,7 +585,7 @@ def test_render_keys_box_and_footer(capsys):
 def test_render_keys_show_id(capsys):
     _wide_stdout()
     output.render_keys([ApiKey(id="1f58376d-7947-9826", name="admin", created_at="2026-06-18T05:14:00Z")], show_id=True)
-    assert "1f58…9826" in capsys.readouterr().out  # short id column when --show-id
+    assert "1f58376d-7947-9826" in capsys.readouterr().out
     output.configure(no_color=True, quiet=False)
 
 
@@ -659,7 +659,7 @@ def test_render_users_box_and_footer(capsys):
     output.users_footer(users)
     cap = capsys.readouterr()
     out, err = cap.out, cap.err
-    assert "users · 3" in out
+    assert "USERS · 3" in out
     for c in ("email", "access", "perms", "joined", "status"):
         assert c in out
     assert "root@test" in out and "admin" in out and "06-25" in out
@@ -695,7 +695,7 @@ def test_render_user_show_card_and_perms(capsys):
     out = capsys.readouterr().out  # show is a read view → stdout
     assert "user" in out and "dev@test" in out and "protected" in out
     assert "access admin" in out and "3 permissions" in out and "active" in out
-    assert "permissions · 3" in out and "keys" in out and "dashboards" in out
+    assert "PERMISSIONS · 3" in out and "keys" in out and "dashboards" in out
     output.configure(no_color=True, quiet=False)
 
 
@@ -706,7 +706,7 @@ def test_render_user_created_green_card(capsys):
     err = capsys.readouterr().err  # write result → stderr chrome
     assert "user created" in err and "new@test" in err
     assert "access read-only" in err and "2 permissions" in err and "active" in err
-    assert "permissions · 2 · read-only" in err  # suffix shows the set
+    assert "PERMISSIONS · 2 · read-only" in err  # suffix shows the set
     output.configure(no_color=True, quiet=False)
 
 
@@ -718,7 +718,7 @@ def test_render_user_updated_diff(capsys):
     err = capsys.readouterr().err
     assert "permissions updated · dev@test" in err
     assert "now 3" in err and "+1 added" in err and "−1 removed" in err
-    assert "permissions · 3" in err          # NEW set size (struck removal not counted)
+    assert "PERMISSIONS · 3" in err          # NEW set size (struck removal not counted)
     assert "+create" in err and "-write" in err  # added chip / removed ghost (no-color prefixes)
     assert "unchanged" in err                # legend
     output.configure(no_color=True, quiet=False)
@@ -1146,7 +1146,7 @@ def test_render_settings_box(capsys):
     output.render_settings(rows)
     cap = capsys.readouterr()
     out, err = cap.out, cap.err
-    assert "settings · 3" in out  # the count glows in the title
+    assert "SETTINGS · 3" in out
     for c in ("key", "value", "type", "updated"):
         assert c in out
     assert "session_ttl_secs" in out and "86400" in out and "integer" in out
@@ -1231,10 +1231,10 @@ def test_render_incidents_box_and_footer(capsys):
     output.incidents_footer(incs)
     cap = capsys.readouterr()
     out, err = cap.out, cap.err
-    assert "issues · 3" in out
+    assert "ISSUES · 3" in out
     for c in ("id", "alert", "severity", "state", "opened", "assignees"):
         assert c in out
-    assert "1f58…9826" in out                       # short id (the handle)
+    assert "1f5803aaaaaabbbbcccc000000009826" in out
     assert "—" in out                               # manual incident (no alert_name) + no assignees
     assert "firing" in out and "acknowledged" in out and "resolved" in out
     assert "3 issues" in err and "1 firing" in err and "1 resolved" in err  # footer distribution
@@ -1271,7 +1271,7 @@ def test_render_incident_show_sections(capsys):
     )
     output.render_incident_show(inc)
     out = capsys.readouterr().out
-    assert "p95 latency" in out and "1f58…9826" in out
+    assert "p95 latency" in out and "1f5803aaaaaabbbbcccc000000009826" in out
     assert "acknowledged by ops@example.com" in out and "assigned to a@example.com" in out
     assert "breach" in out and "1240ms" in out
     assert "comments · 2" in out and "on it" in out and "(deleted)" in out
@@ -1357,7 +1357,7 @@ def test_confirm_incident_resolve_headline(capsys, monkeypatch):
     monkeypatch.setattr(output.typer, "confirm", lambda *a, **k: False)
     assert output.confirm_incident_resolve("1f5803aaaaaabbbbcccc000000009826", "p95 latency") is False
     err = capsys.readouterr().err
-    assert "resolve issue" in err and "1f58…9826" in err and "(p95 latency)" in err and "this closes it" in err
+    assert "resolve issue" in err and "1f5803aaaaaabbbbcccc000000009826" in err and "(p95 latency)" in err and "this closes it" in err
 
 
 def test_incident_plain_feedback_lines(capsys):
@@ -1374,14 +1374,14 @@ def test_incident_plain_feedback_lines(capsys):
     output.incident_comment_not_found("c0ffee001111")
     output.incident_failed("a@x.com is not an operator")
     err = capsys.readouterr().err
-    assert "acknowledged issue 1f58…9826" in err
-    assert "resolved issue 1f58…9826" in err
-    assert "assigned 1f58…9826" in err and "a@example.com, b@example.com" in err
-    assert "cleared assignees on 1f58…9826" in err
+    assert f"acknowledged issue {iid}" in err
+    assert f"resolved issue {iid}" in err
+    assert f"assigned {iid}" in err and "a@example.com, b@example.com" in err
+    assert f"cleared assignees on {iid}" in err
     assert "subscribed you to issue" in err and "unsubscribed x@y.z from issue" in err
     assert "deleted comment" in err
-    assert "no issue 1f58…9826" in err and "fp issues list" in err
-    assert "no comment" in err and "c0ff…1111" in err
+    assert f"no issue {iid}" in err and "fp issues list" in err
+    assert "no comment" in err and "c0ffee001111" in err
     assert "a@x.com is not an operator" in err
 
 
@@ -1399,7 +1399,7 @@ def test_render_agent_health_configured(capsys):
     _wide_stdout()
     output.render_agent_health(configured=True, default_model="claude-x", model_count=3)
     out = capsys.readouterr().out  # data view → stdout
-    assert "assistant" in out and "configured" in out
+    assert "ASSISTANT" in out and "configured" in out
     assert "default model claude-x" in out and "3 models available" in out
     output.configure(no_color=True, quiet=False)
 
@@ -1417,7 +1417,7 @@ def test_render_agent_models_marks_default(capsys):
     _wide_stdout()
     output.render_agent_models(["m-default", "m-fast"], default_model="m-default")
     out = capsys.readouterr().out
-    assert "models · 2" in out and "m-default" in out and "m-fast" in out and "default" in out
+    assert "MODELS · 2" in out and "m-default" in out and "m-fast" in out and "default" in out
     output.configure(no_color=True, quiet=False)
 
 
@@ -1425,7 +1425,7 @@ def test_render_agent_models_empty(capsys):
     _wide_stdout()
     output.render_agent_models([], default_model=None)
     out = capsys.readouterr().out
-    assert "models · 0" in out and "no models reported" in out
+    assert "MODELS · 0" in out and "no models reported" in out
     output.configure(no_color=True, quiet=False)
 
 
@@ -1438,12 +1438,12 @@ def test_render_agent_chats_box(capsys):
     output.render_agent_chats(chats)
     cap = capsys.readouterr()
     out, err = cap.out, cap.err
-    assert "chats · 2" in out  # the count glows in the title
+    assert "CHATS · 2" in out
     for c in ("chat-id", "title", "messages", "updated"):
         assert c in out
     assert "older" in out and "newer" in out
     assert "07854990" in out and "17bf35c3" in out  # short copy-friendly handle (first 8)
-    assert "dade" not in out                          # the rest of the id is not shown
+    assert "07854990-dade-4dea-aaaa" in out
     assert out.index("newer") < out.index("older")   # newest first
     assert "open one with" not in err                 # the footer hint was removed
     output.configure(no_color=True, quiet=False)
@@ -1492,7 +1492,7 @@ def test_render_agent_delete_preview_and_feedback(capsys):
     output.print_cancelled("nothing deleted")  # boxed cancel, like the command uses
     err = capsys.readouterr().err
     assert "delete chat" in err and "perf review" in err and "4 messages" in err
-    assert "07854990" in err and "dade" not in err  # short id in the preview
+    assert "07854990-dade" in err  # full id in the preview
     assert "deleted chat" in err and "perf review" in err  # boxed deleted
     assert "cancelled" in err and "nothing deleted" in err  # boxed cancel
     output.configure(no_color=True, quiet=False)
@@ -1517,7 +1517,7 @@ def test_agent_ask_chrome_lines(capsys):
     output.agent_unconfigured_note()
     err = capsys.readouterr().err
     assert "used tool: run_query" in err
-    assert "new chat" in err and 'fp agent ask --chat 07854990 "…"' in err  # short id, positional msg
+    assert "new chat" in err and 'fp agent ask --chat 07854990-dade-4dea "…"' in err
     assert "assistant error: boom" in err
     assert "chat not found" in err and "fp agent chats" in err  # boxed not-found
     assert "isn't configured" in err and "fp agent health" in err
@@ -1642,7 +1642,7 @@ def test_policy_list_counts_policies_and_captions_versions(capsys):
     _wide_stdout()
     output.render_policies([pv("a", 1), pv("b", 2), pv("b", 1)])
     out = capsys.readouterr().out
-    assert "policies · 2 · 3 versions" in out
+    assert "POLICIES · 2 · 3 versions" in out
     # newest version of each policy first, rather than server order
     b_rows = [ln for ln in out.splitlines() if "b" in ln and "v" in ln
               and ("v1" in ln or "v2" in ln)]
@@ -1657,7 +1657,7 @@ def test_policy_list_omits_the_caption_when_each_policy_has_one_version(capsys):
                                           source=None, created_at="", created_by=None,
                                           disabled=False, archived=False)])
     out = capsys.readouterr().out
-    assert "policies · 1" in out and "versions" not in out
+    assert "POLICIES · 1" in out and "versions" not in out
 
 
 # ─────────────────────────────────────────────────────────────────────────────

package/.next/standalone/fp-cloud-cli/tests/test_round4_behaviour.py

@@ -0,0 +1,144 @@
+import json
+from io import StringIO
+
+import pytest
+import respx
+from rich.console import Console
+
+from fp_cli import output
+from fp_cli.app import app
+
+BASE = "http://dash.test"
+SOURCE = 'EvalResult(score=Score(0.7), reasoning="Tool usage")'
+
+
+@pytest.mark.parametrize("command", ["create", "deploy"])
+@pytest.mark.parametrize("force", [False, True])
+@pytest.mark.parametrize("json_mode", [False, True])
+@pytest.mark.parametrize("width", [80, 140])
+def test_failed_preview_stops_unless_force(logged_in, runner, tmp_path, monkeypatch, command, force, json_mode, width):
+    monkeypatch.setenv("COLUMNS", str(width))
+    file = tmp_path / "source.py"
+    file.write_text(SOURCE)
+    old = {"id": "old", "eval_key": "tool_usage", "eval_version": "1", "enabled": True}
+    preview = {
+        "runs": [
+            {"session_id": "run-check", "ok": True,
+             "result": {"score": {"value": 0.7}, "reasoning": "Seven tool calls"}},
+            {"session_id": "failed-check", "ok": False, "error_kind": "invalid_result",
+             "error_message": "Missing result score"},
+        ],
+        "summary": {"succeeded": 1, "failed": 1, "skipped": 0},
+    }
+    capture = StringIO()
+    original_configure = output.configure
+
+    def configure_capture(**kwargs):
+        original_configure(**kwargs)
+        if not json_mode:
+            output._stdout = Console(file=capture, width=width, no_color=True)
+
+    monkeypatch.setattr(output, "configure", configure_capture)
+    with respx.mock(assert_all_called=False) as mock:
+        mock.get(BASE + "/api/evaluator-definitions").respond(200, json=[old])
+        mock.post(BASE + "/api/evaluator-definitions/dry-run").respond(200, json=preview)
+        created = {
+            "id": "new", "eval_key": "tool_usage", "eval_version": "2", "enabled": True,
+        }
+        create = mock.post(BASE + "/api/evaluator-definitions")
+        patch = mock.patch(BASE + "/api/evaluator-definitions/old").respond(200, json={})
+
+        def preview_happened_first(request):
+            # Data must already be visible to the user when creation starts.
+            if not json_mode:
+                assert "Missing result score" in capture.getvalue()
+                assert "Seven tool calls" in capture.getvalue()
+            import httpx
+            return httpx.Response(201, json=created)
+
+        create.mock(side_effect=preview_happened_first)
+        flags = ["--key", "tool_usage", "--name", "Tool usage"] if command == "create" else ["tool_usage"]
+        result = runner.invoke(app, [
+            *(["--json"] if json_mode else []), "evaluators", command, *flags,
+            "--code", str(file), "--dry-run-on", "run-check", "--dry-run-on", "failed-check",
+            *(["--force"] if force else []),
+        ])
+    assert result.exit_code == (0 if force else 1), result.output
+    assert create.called is force
+    assert patch.called is (force and command == "deploy")
+    if json_mode:
+        payload = json.loads(result.stdout)
+        if force:
+            assert payload == (
+                {"created": {"id": "new", "eval_key": "tool_usage", "eval_version": "2", "enabled": True}, "disabled": ["old"]}
+                if command == "deploy" else {"id": "new", "eval_key": "tool_usage", "eval_version": "2", "enabled": True}
+            )
+        else:
+            assert payload["exit_code"] == 1
+            assert "Missing result score" in payload["error"]
+    else:
+        rendered = capture.getvalue()
+        assert "failed-check · failed" in rendered
+        assert "Missing result score" in rendered
+        if force:
+            assert "--force" in result.stderr
+            assert rendered.index("Missing result score") < rendered.index(
+                "deployed tool_usage" if command == "deploy" else "EVALUATOR CREATED")
+
+
+@pytest.mark.parametrize("width", [80, 140])
+def test_preview_skip_has_deliberate_outcome(logged_in, runner, tmp_path, monkeypatch, width):
+    monkeypatch.setenv("COLUMNS", str(width))
+    file = tmp_path / "source.py"
+    file.write_text(SOURCE)
+    with respx.mock as mock:
+        mock.post(BASE + "/api/evaluator-definitions/dry-run").respond(200, json={
+            "runs": [{"session_id": "run-check", "ok": True, "applicable": False, "result": None}],
+            "summary": {"succeeded": 0, "failed": 0, "skipped": 1},
+        })
+        result = runner.invoke(app, ["evaluators", "dry-run", "--code", str(file), "--session", "run-check"])
+    assert result.exit_code == 0, result.output
+    assert "○ run-check · skipped" in result.stdout
+
+
+@pytest.mark.parametrize("key_mode", [False, True])
+def test_deploy_file_without_key(logged_in, runner, tmp_path, key_mode):
+    file = tmp_path / "spec.json"
+    file.write_text(json.dumps({"display_name": "Tool usage", "evaluator_source": SOURCE, "eval_version": "2"}))
+    prefix = "/v1" if key_mode else "/api"
+    with respx.mock as mock:
+        mock.get(BASE + prefix + "/evaluator-definitions").respond(200, json=[])
+        create = mock.post(BASE + prefix + "/evaluator-definitions").respond(201, json={"id": "new"})
+        result = runner.invoke(app, [
+            "--json", *(["--api-key", "key"] if key_mode else []),
+            "evaluators", "deploy", "tool_usage", "--file", str(file),
+        ])
+    assert result.exit_code == 0, result.output
+    assert json.loads(create.calls[0].request.content)["eval_key"] == "tool_usage"
+    assert json.loads(result.stdout) == {"created": {"id": "new"}, "disabled": []}
+
+
+@pytest.mark.parametrize("command", [["plan"], ["plan", "claim", "team_offer"]])
+@pytest.mark.parametrize("json_mode", [False, True])
+def test_plan_needs_signed_in_person_before_http(logged_in, runner, command, json_mode):
+    with respx.mock as mock:
+        result = runner.invoke(app, ["--api-key", "key", *(["--json"] if json_mode else []), *command])
+        assert not mock.calls
+    assert result.exit_code == 2
+    expected = f"`fp {' '.join(command[:2])}` needs a signed-in session: run `fp login`"
+    assert expected in result.output
+
+
+@pytest.mark.parametrize("key_mode", [False, True])
+@pytest.mark.parametrize("body,expected", [
+    ({}, "not available on this FailproofAI Cloud yet"),
+    ({"error": "unknown route"}, "not available on this FailproofAI Cloud yet"),
+    ({"error": "setting not found"}, "setting not found"),
+])
+def test_setting_key_404_on_older_cloud(logged_in, runner, key_mode, body, expected):
+    prefix = "/v1" if key_mode else "/api"
+    with respx.mock as mock:
+        mock.put(BASE + prefix + "/settings/otel.ingest_enabled").respond(404, json=body)
+        result = runner.invoke(app, ["--json", *(["--api-key", "key"] if key_mode else []), "otel", "enable"])
+    assert result.exit_code == 6, result.output
+    assert json.loads(result.stdout)["error"] == expected

package/.next/standalone/fp-cloud-cli/tests/test_screen_language.py

@@ -0,0 +1,80 @@
+import json
+from io import StringIO
+
+from rich.console import Console
+
+from fp_cli import output, theme
+from fp_cli.models import AgentEvent
+
+
+def test_two_accent_palette():
+    assert theme.ACCENT == "#66d1b5"
+    assert theme.PINK == "#e4587d"
+    assert {theme.SUCCESS, theme.ERROR, theme.AMBER, theme.BLUE, theme.SCORE_HIGH} <= {theme.ACCENT, theme.PINK}
+
+
+def test_permission_risk_order_is_independent_of_shared_colours():
+    assert theme.PERM_WRITE == theme.PERM_ACTION == theme.PERM_DANGER
+    assert sorted(["delete", "run", "write", "read"], key=output._perm_act_key) == [
+        "read", "write", "run", "delete",
+    ]
+
+
+def test_open_table_and_header(capsys):
+    output.configure(no_color=True)
+    output.print_table(["Id", "URL"], [["run-001", "https://cloud.example.test"]], title="Resources")
+    text = capsys.readouterr().out
+    assert "failproof ai v" in text
+    assert "RESOURCES" in text
+    assert not any(c in text for c in "╭╮╰╯┌┐└┘─━│┃")
+
+
+def test_legacy_panels_use_open_screen(capsys):
+    output.configure(no_color=True)
+    output.render_events([AgentEvent(id=1, session_id="run-001", agent_id="agent", event_type="tool_use", ts="2026-10-13T00:00:00Z")])
+    text = capsys.readouterr().out
+    assert "EVENTS" in text
+    assert "failproof ai v" in text
+    assert not any(c in text for c in "╭╮╰╯┌┐└┘─━│┃")
+
+
+def test_narrow_ids_and_urls_are_not_cut(capsys):
+    output.configure(no_color=True)
+    output._stdout = Console(width=45, no_color=True)
+    id = "session-001-abcdefghijklmnopqrstuvwxyz-123456789"
+    url = "https://cloud.example.test/organizations/reliability/sessions/" + id
+    output.print_table(["Id", "URL"], [[id, url]])
+    compact = "".join(capsys.readouterr().out.split())
+    assert id in compact and url in compact
+    assert "…" not in compact
+
+
+def test_json_bytes_unchanged(capsys):
+    output.configure(json=True)
+    data = {"session_id": "run-001", "scores": {"quality": 0.9}}
+    output.emit_json(data)
+    assert capsys.readouterr().out == json.dumps(data, indent=2, ensure_ascii=False) + "\n"
+
+
+def test_generated_help_has_no_boxes(logged_in, runner):
+    from fp_cli.app import app
+    result = runner.invoke(app, ["evaluators", "create", "--help"])
+    assert result.exit_code == 0
+    assert "Code:" in result.stdout and "Judge:" in result.stdout and "Jev:" in result.stdout
+    assert not any(c in result.stdout for c in "╭╮╰╯┌┐└┘─━│┃")
+
+
+def test_shared_key_value_labels_are_grey():
+    capture = StringIO()
+    output.configure()
+    output._stdout = Console(file=capture, width=140, force_terminal=True, color_system="truecolor")
+    output.print_table(["Field", "Value"], [["enabled", "true"]])
+    assert "\x1b[38;2;118;127;139menabled" in capture.getvalue()
+
+
+def test_help_typeable_commands_are_pink():
+    capture = StringIO()
+    output.configure()
+    output._stdout = Console(file=capture, width=140, force_terminal=True, color_system="truecolor")
+    output.render_top_level_help()
+    assert "\x1b[38;2;228;88;125mfp login" in capture.getvalue()

package/.next/standalone/fp-cloud-cli/tests/test_session_reevaluate.py

@@ -0,0 +1,33 @@
+import pytest
+import respx
+
+from fp_cli.app import app
+
+
+@pytest.mark.parametrize("key_mode", [False, True])
+@pytest.mark.parametrize("status", [200, 404])
+def test_re_evaluate(logged_in, runner, key_mode, status):
+    prefix = "v1" if key_mode else "api"
+    with respx.mock:
+        route = respx.post(f"http://dash.test/{prefix}/sessions/run-001/re-evaluate").respond(
+            status, json={"queued": True} if status == 200 else {"error": "unknown route"},
+        )
+        result = runner.invoke(app, ["--json", *(["--api-key", "key"] if key_mode else []), "sessions", "re-evaluate", "run-001"])
+    assert route.called
+    assert result.exit_code == (0 if status == 200 else 6), result.output
+    if status == 404:
+        assert "not available on this FailproofAI Cloud yet" in result.stdout
+
+
+@pytest.mark.parametrize("key_mode", [False, True])
+@pytest.mark.parametrize("error_body", [{}, {"error": "session not found"}])
+def test_re_evaluate_missing_session(logged_in, runner, key_mode, error_body):
+    prefix = "v1" if key_mode else "api"
+    with respx.mock as mock:
+        mock.post(f"http://dash.test/{prefix}/sessions/missing/re-evaluate").respond(404, json=error_body)
+        result = runner.invoke(app, [
+            "--json", *(["--api-key", "key"] if key_mode else []), "sessions", "re-evaluate", "missing",
+        ])
+    assert result.exit_code == 6, result.output
+    assert "not found" in result.stdout
+    assert "not available on this FailproofAI Cloud yet" not in result.stdout