test(ixp): grade the value-predicate smokes via cli_called

tests/tasks/uipath-ixp/smoke/create_field.yaml

@@ -10,9 +10,11 @@ run_limits:
   expected_turns: 5
 
 sandbox:
-  mock_path_dirs: [mocks]
-  template_sources:
-    - {type: template_dir, path: ../_shared/mock_template}
+  record_cli:
+    - tool: uip
+      stderr: "uip (mock): not connected to a tenant in this sandbox; offline smoke mock.\n"
+    - tool: curl
+      stderr: "curl (mock): network egress is disabled in this offline smoke sandbox.\n"
 
 initial_prompt: |
   In the IXP project my_invoices-f1afa9ef-ixp, add a new field named "email"
@@ -25,10 +27,15 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
-    description: "Agent added field \"email\" to \"Subscriber information\" via `fields add`"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+fields\s+add\b(?=.*my_invoices-f1afa9ef-ixp)(?=.*--group\s+["']?Subscriber information)(?=.*--field\s+["']?email).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent added the field via `fields add my_invoices-f1afa9ef-ixp --group 'Subscriber information' --field email`, with a non-empty --type and --instructions"
+    verb: "ixp fields add"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp"]
+    flags:
+      group: "Subscriber information"
+      field: "email"
+      type: {matches_regex: ".+"}
+      instructions: {matches_regex: ".+"}
     weight: 3.0
     pass_threshold: 1.0

tests/tasks/uipath-ixp/smoke/delete_multiple_fields.yaml

@@ -12,9 +12,11 @@ run_limits:
   expected_turns: 6
 
 sandbox:
-  mock_path_dirs: [mocks]
-  template_sources:
-    - {type: template_dir, path: ../_shared/mock_template}
+  record_cli:
+    - tool: uip
+      stderr: "uip (mock): not connected to a tenant in this sandbox; offline smoke mock.\n"
+    - tool: curl
+      stderr: "curl (mock): network egress is disabled in this offline smoke sandbox.\n"
 
 initial_prompt: |
   In the IXP project my_invoices-f1afa9ef-ixp, delete the fields "field1" and
@@ -27,18 +29,26 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
-    description: "Agent deleted field1 via `fields delete`"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+fields\s+delete\b(?=.*field1).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent deleted field1 via `fields delete my_invoices-f1afa9ef-ixp --group 'Invoice Header' --field field1 -y/--yes`"
+    verb: "ixp fields delete"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp"]
+    flags:
+      group: "Invoice Header"
+      field: "field1"
+      "yes": {present: true, aliases: ["y"]}
     weight: 2.0
     pass_threshold: 1.0
 
-  - type: file_matches_regex
-    description: "Agent deleted field2 via `fields delete` (both, not just one)"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+fields\s+delete\b(?=.*field2).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent deleted field2 via `fields delete my_invoices-f1afa9ef-ixp --group 'Invoice Header' --field field2 -y/--yes` (both, not just one)"
+    verb: "ixp fields delete"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp"]
+    flags:
+      group: "Invoice Header"
+      field: "field2"
+      "yes": {present: true, aliases: ["y"]}
     weight: 2.0
     pass_threshold: 1.0

tests/tasks/uipath-ixp/smoke/mark_missing.yaml

@@ -10,9 +10,11 @@ run_limits:
   expected_turns: 6
 
 sandbox:
-  mock_path_dirs: [mocks]
-  template_sources:
-    - {type: template_dir, path: ../_shared/mock_template}
+  record_cli:
+    - tool: uip
+      stderr: "uip (mock): not connected to a tenant in this sandbox; offline smoke mock.\n"
+    - tool: curl
+      stderr: "curl (mock): network egress is disabled in this offline smoke sandbox.\n"
 
 initial_prompt: |
   In the IXP project my_invoices-f1afa9ef-ixp, document doc-abc-123, the field
@@ -26,10 +28,23 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
-    description: "Agent marked the field missing via `labellings mark-missing` on the right project + document + field id"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+labellings\s+mark-missing\b(?=.*my_invoices-f1afa9ef-ixp)(?=.*doc-abc-123)(?=.*a7c3e9105f2b4d86).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent marked the field missing via `labellings mark-missing my_invoices-f1afa9ef-ixp doc-abc-123 --fields a7c3e9105f2b4d86`"
+    verb: "ixp labellings mark-missing"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp", "doc-abc-123"]
+    flags:
+      fields: {equals: "a7c3e9105f2b4d86", aliases: ["f"]}
     weight: 3.0
     pass_threshold: 1.0
+
+  - type: cli_called
+    description: "Did NOT mark any field other than a7c3e9105f2b4d86 missing"
+    verb: "ixp labellings mark-missing"
+    tool: "uip"
+    flags:
+      fields: {matches_regex: "^(?!a7c3e9105f2b4d86$)", aliases: ["f"]}
+    min_count: 0
+    max_count: 0
+    weight: 1.5
+    pass_threshold: 1.0