Variable substitution 3

CHANGELOG.md

@@ -7,6 +7,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+- Variable substitution in python tool.
+
 ## [v0.11.5] - 6.11.2025
 
 ### Changed
@@ -60,6 +63,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Use typescript autogen for backend types in frontend.
 - Try to enforce using metric tools rather than downloading assets.
 - Rule to avoid overvalidating.
+- Plot through python + plotly.
+
+## Fixed
+- Images not appearing in Literature search (Prompt refactor).
+- Update uv lock package.
 
 ## [v0.10.0] - 2.10.2025
 

backend/alembic/versions/92694adcad71_message_content_jsonb.py

@@ -0,0 +1,46 @@
+"""message content jsonb
+
+Revision ID: 92694adcad71
+Revises: cd5a73ea91db
+Create Date: 2025-10-08 17:00:43.134455
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "92694adcad71"
+down_revision: Union[str, None] = "cd5a73ea91db"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column(
+        "messages",
+        "content",
+        existing_type=sa.VARCHAR(),
+        type_=postgresql.JSONB(astext_type=sa.Text()),
+        existing_nullable=False,
+        postgresql_using="content::jsonb",
+    )  # Add this parameter
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column(
+        "messages",
+        "content",
+        existing_type=postgresql.JSONB(astext_type=sa.Text()),
+        type_=sa.VARCHAR(),
+        existing_nullable=False,
+        postgresql_using="content::text",
+    )  # Add this parameter
+    # ### end Alembic commands ###

backend/eval/input/variable_substitution/expected_output.md

@@ -0,0 +1 @@
+I've retrieved the section lengths for morphology 7e69531a-c976-4b4c-be44-881e5e3defdd and created a histogram showing the distribution. The plot displays the frequency of different section lengths across the morphology, which helps visualize the structural characteristics of this neuron.

backend/eval/input/variable_substitution/expected_tool_calls.json

@@ -0,0 +1,22 @@
+[
+  {
+    "name": "obione-morphometrics-getone",
+    "arguments": {
+      "morphology_id": "7e69531a-c976-4b4c-be44-881e5e3defdd",
+      "requested_metrics": ["section_lengths"]
+    }
+  },
+  {
+    "name": "run-python",
+    "arguments": {
+      "python_script": "import plotly.graph_objects as go\n\nfig = go.Figure()\nfig.add_trace(go.Histogram(x=section_lengths, nbinsx=30, name='Section Lengths'))\nfig.update_layout(\n    title='Distribution of Section Lengths',\n    xaxis_title='Section Length (μm)',\n    yaxis_title='Frequency'\n)\nfig.show()\n'plot_generated'",
+      "variables": [
+        {
+          "variable_name": "section_lengths",
+          "tool_call_id": "<morphometrics_tool_call_id>",
+          "jsonpath_reference": "$.*.section_lengths"
+        }
+      ]
+    }
+  }
+]

backend/eval/input/variable_substitution/params.json

@@ -0,0 +1,3 @@
+{
+  "tags": ["morphology", "python", "plotting", "variable_substitution"]
+}

backend/eval/input/variable_substitution/user.md

@@ -0,0 +1 @@
+Get the section_lengths of morphology 7e69531a-c976-4b4c-be44-881e5e3defdd and plot the distribution.

backend/eval/output/scores.json

@@ -1,158 +1,166 @@
 {
-  "total_tests": 19,
+  "total_tests": 20,
   "metrics_df": [
     {
       "test_name": "cerebellum_morphologies",
-      "Correctness [GEval]": 0.49012383446213326,
+      "Correctness [GEval]": 0.38821394911329454,
       "Tool Correctness": 1.0,
-      "Argument Correctness": 0.0,
+      "Argument Correctness": 0.5,
       "Deterministic Argument Correctness": 0.5,
       "Overall Argument Correctness": 0.5
     },
     {
       "test_name": "connectivity_metrics",
-      "Correctness [GEval]": 0.7718103598384992,
-      "Tool Correctness": 1.0,
-      "Argument Correctness": 0.5,
-      "Deterministic Argument Correctness": 1.0,
-      "Overall Argument Correctness": 1.0
+      "Correctness [GEval]": 0.29090204872481407,
+      "Tool Correctness": 0.5,
+      "Argument Correctness": 0.0,
+      "Deterministic Argument Correctness": 0.5,
+      "Overall Argument Correctness": 0.5
     },
     {
       "test_name": "connectivity_metrics_extra_filters",
-      "Correctness [GEval]": 0.792225826850805,
-      "Tool Correctness": 1.0,
-      "Argument Correctness": 0.5,
-      "Deterministic Argument Correctness": 1.0,
-      "Overall Argument Correctness": 1.0
+      "Correctness [GEval]": 0.29917647472477826,
+      "Tool Correctness": 0.5,
+      "Argument Correctness": 0.0,
+      "Deterministic Argument Correctness": 0.5,
+      "Overall Argument Correctness": 0.5
     },
     {
       "test_name": "get_specific_circuit",
-      "Correctness [GEval]": 0.8044122798124238,
+      "Correctness [GEval]": 0.786330785728382,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "ion_channel",
-      "Correctness [GEval]": 0.547455303051177,
+      "Correctness [GEval]": 0.6100691296561099,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
-      "Deterministic Argument Correctness": 0.0,
+      "Deterministic Argument Correctness": 0.5,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "ion_channel_recording",
-      "Correctness [GEval]": 0.39644459702330714,
+      "Correctness [GEval]": 0.38548079269064617,
       "Tool Correctness": 1.0,
-      "Argument Correctness": 1.0,
+      "Argument Correctness": 0.5,
       "Deterministic Argument Correctness": 0.75,
-      "Overall Argument Correctness": 1.0
+      "Overall Argument Correctness": 0.75
     },
     {
       "test_name": "morphology_studies",
-      "Correctness [GEval]": 0.5144470520430822,
+      "Correctness [GEval]": 0.7292190326367721,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 0.3333333333333333,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "neuroscientists_search",
-      "Correctness [GEval]": 0.4457554724314045,
+      "Correctness [GEval]": 0.23270807954872502,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 0.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "platform_explore",
-      "Correctness [GEval]": 0.6225318559484088,
+      "Correctness [GEval]": 0.5105671034963964,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "platform_news",
-      "Correctness [GEval]": 0.6677989147190616,
+      "Correctness [GEval]": 0.6191804216904908,
       "Tool Correctness": 1.0,
       "Argument Correctness": 0.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "platform_ui_simulate",
-      "Correctness [GEval]": 0.4022727821989431,
+      "Correctness [GEval]": 0.5028573368554985,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "platform_viewing",
-      "Correctness [GEval]": 0.5599305388335759,
+      "Correctness [GEval]": 0.6354826972531142,
       "Tool Correctness": 1.0,
       "Argument Correctness": 0.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "plotting",
-      "Correctness [GEval]": 0.5154697882424646,
+      "Correctness [GEval]": 0.6565033805474136,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 0.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "simulation_tutorial",
-      "Correctness [GEval]": 0.3643381261146244,
-      "Tool Correctness": 0.0,
-      "Argument Correctness": 1.0,
-      "Deterministic Argument Correctness": 0.0,
-      "Overall Argument Correctness": 1.0
+      "Correctness [GEval]": 0.35381328035923715,
+      "Tool Correctness": 1.0,
+      "Argument Correctness": 0.0,
+      "Deterministic Argument Correctness": 0.5,
+      "Overall Argument Correctness": 0.5
     },
     {
       "test_name": "sin_plot",
-      "Correctness [GEval]": 0.7090979511047657,
+      "Correctness [GEval]": 0.7336164984672461,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 0.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "software_docs_entitysdk",
-      "Correctness [GEval]": 0.4988197224981158,
-      "Tool Correctness": 0.0,
+      "Correctness [GEval]": 0.4779583628701857,
+      "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
-      "Deterministic Argument Correctness": 0.0,
+      "Deterministic Argument Correctness": 0.6666666666666666,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "software_docs_obione",
-      "Correctness [GEval]": 0.5545155228481053,
+      "Correctness [GEval]": 0.37049372351787807,
       "Tool Correctness": 1.0,
       "Argument Correctness": 0.0,
-      "Deterministic Argument Correctness": 0.3333333333333333,
-      "Overall Argument Correctness": 0.3333333333333333
+      "Deterministic Argument Correctness": 0.6666666666666666,
+      "Overall Argument Correctness": 0.6666666666666666
     },
     {
       "test_name": "species_list",
-      "Correctness [GEval]": 0.5208178183484194,
+      "Correctness [GEval]": 0.5602827980475861,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 1.0,
       "Overall Argument Correctness": 1.0
     },
     {
       "test_name": "thalamus_id",
-      "Correctness [GEval]": 0.46545896861337094,
+      "Correctness [GEval]": 0.525632830123433,
       "Tool Correctness": 1.0,
       "Argument Correctness": 1.0,
       "Deterministic Argument Correctness": 0.0,
       "Overall Argument Correctness": 1.0
+    },
+    {
+      "test_name": "variable_substitution",
+      "Correctness [GEval]": 0.31259073486638117,
+      "Tool Correctness": 0.5,
+      "Argument Correctness": 0.75,
+      "Deterministic Argument Correctness": 0.5,
+      "Overall Argument Correctness": 0.75
     }
   ],
-  "created_at": "2025-11-06 11:18:54.068665"
-}
+  "created_at": "2025-11-14 18:19:53.061708"
+}