feat: add 7 evaluator factories with tests and updated dependencies (#13)

Author: zurfjereluhmie

Makefile

@@ -1,6 +1,6 @@
 .DEFAULT_GOAL := help
 
-.PHONY: help install install-dev install-demo lint format typecheck test check fix pre-commit docs docs-serve clean
+.PHONY: help install install-dev install-demo sync lint format typecheck test check fix pre-commit docs docs-serve clean
 
 help: ## Show this help message
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) \
@@ -15,6 +15,9 @@ install-dev: ## Install package with dev dependencies
 install-demo: ## Install package with dev + demo dependencies
 	uv sync --group dev --group demo
 
+sync: ## Install all extras and dependency groups (for local development)
+	uv sync --all-extras --all-groups
+
 lint: ## Run ruff linter (check only)
 	uv run ruff check src tests demo
 

pyproject.toml

@@ -14,14 +14,16 @@ dependencies = [
     "arize-phoenix>=13.0,<14",
     "pandas>=2.0",
     "click>=8.0",
+    "jsonschema>=4.0",
     "tomli>=2.0; python_version < '3.11'",
 ]
 
 [project.optional-dependencies]
-langgraph = ["langgraph>=0.2"]
-langchain = ["openinference-instrumentation-langchain>=0.1"]
-openai    = ["openinference-instrumentation-openai>=0.1"]
-all       = ["evalwire[langgraph,langchain,openai]"]
+langgraph  = ["langgraph==1.1.3"]
+langchain  = ["openinference-instrumentation-langchain>=0.1"]
+openai     = ["openinference-instrumentation-openai>=0.1"]
+llm-judge  = ["langchain-core>=1.0,<2"]
+all        = ["evalwire[langgraph,langchain,openai,llm-judge]"]
 
 [project.scripts]
 evalwire = "evalwire.cli:main"
@@ -48,11 +50,11 @@ dev = [
 demo = [
     "evalwire[langgraph,openai]",
     "arize-phoenix>=13.0,<14",
-    "langchain>=0.3,<0.4",
-    "langchain-core>=0.3,<0.4",
-    "langchain-openai>=0.2,<0.3",
-    "langgraph>=1.0,<2",
-    "openai>=1.0,<2",
+    "langchain>=1.0,<2",
+    "langchain-core>=1.0,<2",
+    "langchain-openai>=1.0,<2",
+    "langgraph==1.1.3",
+    "openai>=2.0,<3",
     "openinference-instrumentation-langchain>=0.1,<0.2",
-    "python-dotenv>=1.0",
+    "python-dotenv>=1.0,<2",
 ]

src/evalwire/__init__.py

@@ -1,14 +1,31 @@
 """evalwire — systematic evaluation of LangGraph nodes with Arize Phoenix."""
 
-from evalwire.evaluators import make_membership_evaluator, make_top_k_evaluator
+from evalwire.evaluators import (
+    make_contains_evaluator,
+    make_exact_match_evaluator,
+    make_json_match_evaluator,
+    make_llm_judge_evaluator,
+    make_membership_evaluator,
+    make_numeric_tolerance_evaluator,
+    make_regex_evaluator,
+    make_schema_evaluator,
+    make_top_k_evaluator,
+)
 from evalwire.observability import setup_observability
 from evalwire.runner import ExperimentRunner
 from evalwire.uploader import DatasetUploader
 
 __all__ = [
     "DatasetUploader",
     "ExperimentRunner",
+    "make_contains_evaluator",
+    "make_exact_match_evaluator",
+    "make_json_match_evaluator",
+    "make_llm_judge_evaluator",
     "make_membership_evaluator",
+    "make_numeric_tolerance_evaluator",
+    "make_regex_evaluator",
+    "make_schema_evaluator",
     "make_top_k_evaluator",
     "setup_observability",
     # LangGraph helpers — available when the `evalwire[langgraph]` extra is