From 75cb56ef336e1c105b4464ce32b9522690959fda Mon Sep 17 00:00:00 2001
From: Agent G <326687097+readyagentsdev@users.noreply.github.com>
Date: Sun, 27 Sep 2026 17:05:02 +0300
Subject: [PATCH] fix(cli-called): report a zero-match max_count: 0 guard as
 unproven

A max_count: 0 guard that matched nothing used the same detail as a guard the
agent respected, so a typo in verb, tool or flags was invisible. Emit a
distinct 'guard unproven' detail for that case. The score does not change.

Closes #122

Signed-off-by: Agent G <326687097+readyagentsdev@users.noreply.github.com>
---
 src/coder_eval/criteria/cli_called.py |  7 ++++++-
 tests/test_cli_called_criterion.py    | 13 ++++++++++++-
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/src/coder_eval/criteria/cli_called.py b/src/coder_eval/criteria/cli_called.py
index 539b4131a..a8f2f9f70 100644
--- a/src/coder_eval/criteria/cli_called.py
+++ b/src/coder_eval/criteria/cli_called.py
@@ -170,7 +170,12 @@ def _check_impl(
             facets.append(f"flags={sorted(criterion.flags)}")
         wanted = ", ".join(facets)
 
-        if score == 1.0:
+        if score == 1.0 and count == 0 and criterion.max_count == 0:
+            details = (
+                f"0 invocation(s) matched ({wanted}); guard unproven: a typo in verb, tool or flags "
+                "would also match nothing"
+            )
+        elif score == 1.0:
             details = f"{count} invocation(s) matched ({wanted}); satisfies {bound}"
         elif not within_lower:
             # A bare count sends the reader to the sandbox; this criterion exists
diff --git a/tests/test_cli_called_criterion.py b/tests/test_cli_called_criterion.py
index 8667b423e..4de0f96b9 100644
--- a/tests/test_cli_called_criterion.py
+++ b/tests/test_cli_called_criterion.py
@@ -237,7 +237,18 @@ def test_max_count_zero_is_the_negative_guard(self, sandbox_with_log):
             min_count=0,
             max_count=0,
         )
-        assert SuccessChecker(sandbox).check(forbidden).score == 1.0
+        result = SuccessChecker(sandbox).check(forbidden)
+        assert result.score == 1.0
+        assert "guard unproven" in (result.details or "")
+        assert "satisfies" not in (result.details or "")
+
+    def test_positive_match_keeps_satisfies_wording(self, sandbox_with_log):
+        sandbox, sandbox_dir = sandbox_with_log
+        _write_log(sandbox_dir, [_call(["ixp", "fields", "rename", "proj-1"])])
+        criterion = CliCalledCriterion(description="renamed the field", log=LOG, verb="ixp fields rename")
+        result = SuccessChecker(sandbox).check(criterion)
+        assert result.score == 1.0
+        assert result.details == "1 invocation(s) matched (verb='ixp fields rename'); satisfies min_count=1"
 
     def test_max_count_zero_fails_when_called(self, sandbox_with_log):
         sandbox, sandbox_dir = sandbox_with_log