Fix metric units in CLI eval output and template field names

agents/template/agent.py

@@ -31,16 +31,16 @@ def generate(spec: dict) -> bytes:
 
     Args:
         spec: Problem specification dict. Key fields:
-            spec["constraints"]["load_n"]               — load in Newtons
-            spec["constraints"]["load_point_mm"]        — [x, y, z] load point
-            spec["constraints"]["build_volume_mm"]      — [x, y, z] bounding box
-            spec["constraints"]["bolt_pattern_mm"]      — [[y, z], ...] bolt centers
+            spec["constraints"]["load_newtons"]               — load in Newtons
+            spec["constraints"]["load_point_mm"]              — [x, y, z] load point
+            spec["constraints"]["build_volume_mm"]            — [x, y, z] bounding box
+            spec["constraints"]["bolt_pattern_mm"]            — [[y, z], ...] bolt centers
             spec["constraints"]["bolt_diameter_clearance_mm"] — hole clearance
-            spec["constraints"]["min_wall_thickness_mm"] — minimum wall
-            spec["constraints"]["max_overhang_deg"]     — max printable overhang
-            spec["material"]                            — material name
-            spec["safety_factor"]                       — FEA stress safety factor
-            spec["scoring"]["metric"]                   — "mass_grams" | "volume_mm3" | ...
+            spec["constraints"]["min_wall_thickness_mm"]      — minimum wall
+            spec["constraints"]["max_overhang_deg"]           — max printable overhang
+            spec["constraints"]["safety_factor"]              — FEA stress safety factor
+            spec["material"]                                  — material name
+            spec["scoring"]["metric"]                         — "mass_grams" | "stiffness_to_weight" | "deflection_mm"
 
     Returns:
         STEP file as raw bytes (AP214IS schema required).
@@ -53,7 +53,7 @@ def generate(spec: dict) -> bytes:
     constraints = spec["constraints"]
 
     # TODO: read the constraints you need
-    # load_n = constraints["load_n"]
+    # load_n = constraints["load_newtons"]
     # load_point = constraints["load_point_mm"]      # [x, y, z]
     # build_vol = constraints["build_volume_mm"]     # [x, y, z] max box
     # bolt_pattern = constraints["bolt_pattern_mm"]  # [[y, z], ...]

cli.py

@@ -615,7 +615,8 @@ def _run_evaluate(agent_path: str, spec_path: str, verbose: bool) -> dict:
             stress = result.get("fea_stress_mpa", "?")
             allowable = result.get("fea_allowable_mpa", "?")
             elapsed = result.get("elapsed_seconds", "?")
-            score_str = f"{score:.2f}g" if isinstance(score, (int, float)) else str(score)
+            metric = result.get("score_metric", "mass_grams")
+            score_str = _fmt_score(score, metric) if isinstance(score, (int, float)) else str(score)
             elapsed_str = f"{elapsed:.1f}s" if isinstance(elapsed, (int, float)) else str(elapsed)
             _ok(f"score={score_str}  stress={stress}/{allowable} MPa  t={elapsed_str}")
         else:
@@ -635,7 +636,8 @@ def _print_summary_table(results: list[dict]) -> None:
         spec = r.get("spec", "?")
         passed = r.get("passed", False)
         status_str = f"{GREEN}PASS{RESET}" if passed else f"{RED}FAIL{RESET}"
-        score = f"{r['score']:.2f}g" if passed and r.get("score") else "—"
+        metric = r.get("score_metric", "mass_grams")
+        score = _fmt_score(r["score"], metric) if passed and r.get("score") else "—"
         reason = "" if passed else r.get("reason", "")[:30]
         print(f"  {spec:<20} {status_str}     {score:>8}  {reason}")
     print(f"{'─' * 64}\n")