From 1e20ab641c4e587fecf84ffda0fe30ebd10b4a08 Mon Sep 17 00:00:00 2001
From: n-papaioannou <n.papaioannou@ime.life>
Date: Tue, 28 Apr 2026 16:06:32 +0200
Subject: [PATCH] fix(cli): clarify run summary so "failed" stops meaning two
 different things
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The post-run summary line said "{n} out of {total} tests have failed",
but {n} counted only tests with insufficient_evidence. The per-category
bars on the same screen marked tests as failed when they didn't pass
their score threshold. Same word, two definitions, two visibly
contradictory counts (e.g. bars: 8 ✗, line: 6 / 8 failed).

Reword to make the distinction explicit:
- success branch: "All {N} tests produced sufficient evidence to be scored."
- failure branch: "{n} out of {total} tests had insufficient evidence to
  be scored ({ids}). The remaining tests were scored but may still be
  below threshold -- see the per-category bars above."
---
 ifixai/cli/orchestrator.py | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/ifixai/cli/orchestrator.py b/ifixai/cli/orchestrator.py
index 43cf6c5d..2f966401 100644
--- a/ifixai/cli/orchestrator.py
+++ b/ifixai/cli/orchestrator.py
@@ -185,14 +185,21 @@ def _print_insufficient_evidence_summary(result: TestRunResult) -> None:
     insufficient = [br for br in result.test_results if br.insufficient_evidence]
     total = len(result.test_results)
     if not insufficient:
-        click.echo(click.style(f"0 out of {total} tests have failed.", fg="green"))
+        click.echo(
+            click.style(
+                f"All {total} tests produced sufficient evidence to be scored.",
+                fg="green",
+            )
+        )
         return
     inspection_ids = ", ".join(sorted(br.test_id for br in insufficient))
     click.echo(
         click.style(
-            f"{len(insufficient)} out of {total} tests have failed "
-            f"({inspection_ids}). Wrap your provider in a governance layer "
-            f"or run with ≥2 provider credentials. See docs/methodology.md.",
+            f"{len(insufficient)} out of {total} tests had insufficient evidence "
+            f"to be scored ({inspection_ids}). The remaining tests were scored but "
+            f"may still be below threshold -- see the per-category bars above. "
+            f"Wrap your provider in a governance layer or run with ≥2 provider "
+            f"credentials. See docs/methodology.md.",
             fg="yellow",
         )
     )