PyPI - fluxloop-cli - Versions diffs - 0.2.35__tar.gz → 0.3.0__tar.gz - Mend

fluxloop-cli 0.2.35tar.gz → 0.3.0tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (78) hide show

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fluxloop-cli
-Version: 0.2.35
+Version: 0.3.0
 Summary: FluxLoop CLI for running agent simulations
 Author-email: FluxLoop Team <team@fluxloop.dev>
 License: Apache-2.0

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/fluxloop_cli/__init__.py RENAMED Viewed

@@ -2,7 +2,7 @@
 FluxLoop CLI - Command-line interface for running agent simulations.
 """
-__version__ = "0.2.35"
+__version__ = "0.3.0"
 from .main import app

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/fluxloop_cli/evaluation/engine/reporting/html.py RENAMED Viewed

@@ -93,7 +93,7 @@ DEFAULT_TEMPLATE = r"""<!DOCTYPE html>
           </div>
         </div>
         <a
-          href="https://fluxloop.io/"
+          href="https://fluxloop.ai/"
           class="inline-flex items-center gap-2 self-start rounded-full border border-sky-400 px-4 py-2 text-sm font-semibold text-sky-200 transition hover:bg-sky-500/10 focus:outline-none focus:ring-2 focus:ring-sky-400 focus:ring-offset-2 focus:ring-offset-slate-950"
           target="_blank"
           rel="noopener noreferrer"

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/fluxloop_cli/evaluation/report/renderer.py RENAMED Viewed

@@ -13,6 +13,24 @@ from jinja2 import Environment, FileSystemLoader
 logger = logging.getLogger(__name__)
+DEFAULT_METRIC_THRESHOLDS = {
+    "task_completion": {"good": 80, "fair": 60},
+    "hallucination": {"good": 5, "fair": 15},
+    "relevance": {"good": 90, "fair": 80},
+    "tool_usage": {"good": 90, "fair": 80},
+    "user_satisfaction": {"good": 70, "fair": 50},
+    "clarity": {"good": 90, "fair": 80},
+    "persona_consistency": {"good": 85, "fair": 70},
+}
+DISPLAY_TO_CONFIG_METRIC_KEYS = {"tool_usage": "tool_usage_appropriateness"}
+CONFIG_TO_DISPLAY_METRIC_KEYS = {
+    config_key: display_key for display_key, config_key in DISPLAY_TO_CONFIG_METRIC_KEYS.items()
+}
+GENERIC_THRESHOLD_FALLBACK = {"good": 80, "fair": 60}
+INVERTED_METRICS = {"hallucination"}
 class ReportRenderer:
     """Prepares data and renders the HTML report."""
@@ -155,27 +173,56 @@ class ReportRenderer:
         }
     def _transform_eval_criteria(self, eval_config: Dict[str, Any]) -> Dict[str, Any]:
-        metrics = eval_config.get("metrics", {})
-        efficiency = eval_config.get("efficiency", {})
-        key_map = {"tool_usage_appropriateness": "tool_usage"}
-        thresholds = {}
-        for k, v in metrics.items():
-            if "thresholds" in v:
-                thresh = v["thresholds"]
-                good, fair = thresh.get("good", 80), thresh.get("fair", 60)
-                norm_key = key_map.get(k, k)
-                # Inverted logic handling for display strings
-                if k == "hallucination":
-                    thresholds[norm_key] = {
-                        "good": f"<={good}%", "fair": f"{good}-{fair}%", "poor": f">{fair}%"
-                    }
-                else:
-                    thresholds[norm_key] = {
-                        "good": f">={good}%", "fair": f"{fair}-{good}%", "poor": f"<{fair}%"
-                    }
+        metrics_cfg = eval_config.get("metrics", {}) or {}
+        efficiency = eval_config.get("efficiency", {}) or {}
+        thresholds: Dict[str, Dict[str, str]] = {}
+        def _coerce_number(value: Any, fallback: float) -> float:
+            if isinstance(value, (int, float)):
+                return float(value)
+            try:
+                return float(value)
+            except (TypeError, ValueError):
+                return float(fallback)
+        def _format_number(value: float) -> str:
+            if value.is_integer():
+                return str(int(value))
+            return str(value)
+        def _register_threshold(display_key: str, good_val: float, fair_val: float) -> None:
+            good_text, fair_text = _format_number(good_val), _format_number(fair_val)
+            if display_key in INVERTED_METRICS:
+                thresholds[display_key] = {
+                    "good": f"<={good_text}%",
+                    "fair": f"{good_text}-{fair_text}%",
+                    "poor": f">{fair_text}%",
+                }
+            else:
+                thresholds[display_key] = {
+                    "good": f">={good_text}%",
+                    "fair": f"{fair_text}-{good_text}%",
+                    "poor": f"<{fair_text}%",
+                }
+        def _resolve_thresholds(display_key: str, metric_cfg: Dict[str, Any]) -> None:
+            defaults = DEFAULT_METRIC_THRESHOLDS.get(display_key, GENERIC_THRESHOLD_FALLBACK)
+            thresholds_cfg = metric_cfg.get("thresholds", {})
+            good_val = _coerce_number(thresholds_cfg.get("good"), defaults["good"])
+            fair_val = _coerce_number(thresholds_cfg.get("fair"), defaults["fair"])
+            _register_threshold(display_key, good_val, fair_val)
+        for cfg_key, cfg_value in metrics_cfg.items():
+            display_key = CONFIG_TO_DISPLAY_METRIC_KEYS.get(cfg_key, cfg_key)
+            _resolve_thresholds(display_key, cfg_value)
+        for display_key, defaults in DEFAULT_METRIC_THRESHOLDS.items():
+            if display_key not in thresholds:
+                _register_threshold(
+                    display_key,
+                    float(defaults["good"]),
+                    float(defaults["fair"]),
+                )
         eff_flags = {}
         if "output_tokens" in efficiency:

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/fluxloop_cli.egg-info/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fluxloop-cli
-Version: 0.2.35
+Version: 0.3.0
 Summary: FluxLoop CLI for running agent simulations
 Author-email: FluxLoop Team <team@fluxloop.dev>
 License: Apache-2.0

{fluxloop_cli-0.2.35 → fluxloop_cli-0.3.0}/pyproject.toml RENAMED Viewed

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "fluxloop-cli"
-version = "0.2.35"
+version = "0.3.0"
 description = "FluxLoop CLI for running agent simulations"
 authors = [
     {name = "FluxLoop Team", email = "team@fluxloop.dev"}