diff --git a/tests/tasks/uipath-ixp/_shared/mock_template/README.md b/tests/tasks/uipath-ixp/_shared/mock_template/README.md
index 361664d464..f8e8ca9ed4 100644
--- a/tests/tasks/uipath-ixp/_shared/mock_template/README.md
+++ b/tests/tasks/uipath-ixp/_shared/mock_template/README.md
@@ -85,14 +85,22 @@ sinks in the same format, then delegates unchanged to the real CLI.
 Tasks graded with `cli_called` declare `sandbox.record_cli` instead, and the
 framework generates the recorders, seeds the log and PATH-prepends `cli_mocks/`
 itself — no `mock_path_dirs`, no `template_sources`, no `log:` on the criterion.
-Prefer that for a new structured task. This template remains the answer whenever
-a recorder is not enough: `file_matches_regex` over the flat `calls.log`, a read
-that must return fixture data (`mock_template_taxonomy`), or per-invocation
-responses, none of which `record_cli` serves. `calls.jsonl` here has no consumer
-today for that reason — it is the structured sink for tasks that need this
+Prefer that for a new structured task. Since coder-eval 0.12.1 a `record_cli`
+entry can also carry `responses:` — per-invocation canned answers matched by the
+same facets `cli_called` grades with — so serving a read before the graded write
+no longer requires this template. No task uses that yet: a shape-only smoke task
+grades the invocation, not the answer, so the blanket failure below is enough for
+it. Reach for `responses:` when a criterion asserts a value the agent could only
+have CARRIED from a read. This template remains the answer whenever a recorder is
+still not enough: `file_matches_regex` over the flat `calls.log`, or a served read
+that must be STATEFUL, where the answer depends on what the agent already wrote
+(`mock_template_taxonomy`) — `select_rule` is stateless by design. `calls.jsonl`
+here has no consumer today — it is the structured sink for tasks that need this
 template's other behavior too.
 
 A task whose correct path reads before it writes cannot be graded on this mock
 alone — the read fails, so there is nothing to carry into the graded write, and
 an agent that declines to invent the values correctly stops. Overlay
-`mock_template_taxonomy` (listed second) to serve that read.
+`mock_template_taxonomy` (listed second) to serve that read, or — for a task on
+`record_cli` whose grading needs the carried value — declare the read as a
+`responses:` rule on the entry itself.
diff --git a/tests/tasks/uipath-ixp/smoke/list_documents.yaml b/tests/tasks/uipath-ixp/smoke/list_documents.yaml
index c1c3dd6419..efe2368146 100644
--- a/tests/tasks/uipath-ixp/smoke/list_documents.yaml
+++ b/tests/tasks/uipath-ixp/smoke/list_documents.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, list all documents.
@@ -24,10 +26,11 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
-    description: "Agent listed documents via `documents list <Name>`"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+documents\s+list\b(?=.*my_invoices-f1afa9ef-ixp).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent listed documents via `documents list my_invoices-f1afa9ef-ixp`"
+    verb: "ixp documents list"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp"]
+    value_flags: ["limit", "l", "offset", "output"]
     weight: 3.0
     pass_threshold: 1.0
diff --git a/tests/tasks/uipath-ixp/smoke/list_projects.yaml b/tests/tasks/uipath-ixp/smoke/list_projects.yaml
index 3b1445b068..723b3c2a36 100644
--- a/tests/tasks/uipath-ixp/smoke/list_projects.yaml
+++ b/tests/tasks/uipath-ixp/smoke/list_projects.yaml
@@ -11,9 +11,11 @@ run_limits:
   expected_turns: 4
 
 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: |
   A document automation engineer wants to see every IXP project on the
@@ -27,10 +29,9 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
+  - type: cli_called
     description: "Agent invoked `uip ixp projects list`"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+projects\s+list(\s|$))[^\r\n]*$
+    verb: "ixp projects list"
+    tool: "uip"
     weight: 2.0
     pass_threshold: 1.0
diff --git a/tests/tasks/uipath-ixp/smoke/show_predictions.yaml b/tests/tasks/uipath-ixp/smoke/show_predictions.yaml
index da433950a3..8beb40149f 100644
--- a/tests/tasks/uipath-ixp/smoke/show_predictions.yaml
+++ b/tests/tasks/uipath-ixp/smoke/show_predictions.yaml
@@ -11,9 +11,11 @@ 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, show the values the model
@@ -26,10 +28,10 @@ initial_prompt: |
   - Do NOT prompt the user for confirmation — this is an automated test.
 
 success_criteria:
-  - type: file_matches_regex
-    description: "Agent fetched predictions via `labellings get-predictions … doc-abc-123`"
-    path: "mocks/calls.log"
-    pattern: >-
-      (?m)^(?:uip\s+ixp\s+labellings\s+get-predictions\b(?=.*doc-abc-123).*)[^\r\n]*$
+  - type: cli_called
+    description: "Agent fetched predictions via `labellings get-predictions my_invoices-f1afa9ef-ixp doc-abc-123`"
+    verb: "ixp labellings get-predictions"
+    tool: "uip"
+    positional: ["my_invoices-f1afa9ef-ixp", "doc-abc-123"]
     weight: 3.0
     pass_threshold: 1.0