feat(rules): refactor code to rename evaluators to rules

.githooks/pre-push

@@ -16,9 +16,9 @@ echo "pre-push: running make typecheck"
 make typecheck
 
 # Check extras if they exist and have changes
-if [ -d "evaluators/contrib/galileo" ]; then
-  echo "pre-push: checking evaluators/contrib/galileo"
-  cd evaluators/contrib/galileo
+if [ -d "rules/contrib/galileo" ]; then
+  echo "pre-push: checking rules/contrib/galileo"
+  cd rules/contrib/galileo
   uv run --extra dev ruff check --config ../../../pyproject.toml src/
   uv run --extra dev mypy --config-file ../../../pyproject.toml src/
   cd "$REPO_ROOT"

.github/workflows/ci.yml

@@ -74,7 +74,7 @@ jobs:
               "coverage-server.xml",
               "coverage-sdk.xml",
           ]
-          contrib_reports = sorted(path.name for path in Path(".").glob("coverage-evaluators-*.xml"))
+          contrib_reports = sorted(path.name for path in Path(".").glob("coverage-rules-*.xml"))
           print(f"files={','.join([*base_reports, *contrib_reports])}")
           PY
 

.github/workflows/pr-title.yaml

@@ -33,7 +33,7 @@ jobs:
             server
             models
             engine
-            evaluators
+            rules
             ci
             docs
             infra

.github/workflows/release.yaml

@@ -53,23 +53,23 @@ jobs:
         run: |
           rm -rf release-dists
           mkdir -p release-dists/models
-          mkdir -p release-dists/evaluators/builtin
-          mkdir -p release-dists/evaluators/contrib
-          mkdir -p release-dists/pypi/evaluators
+          mkdir -p release-dists/rules/builtin
+          mkdir -p release-dists/rules/contrib
+          mkdir -p release-dists/pypi/rules
           mkdir -p release-dists/sdks/python
           mkdir -p release-dists/server
 
           cp -R models/dist release-dists/models/
-          cp -R evaluators/builtin/dist release-dists/evaluators/builtin/
-          cp evaluators/builtin/dist/* release-dists/pypi/evaluators/
+          cp -R rules/builtin/dist release-dists/rules/builtin/
+          cp rules/builtin/dist/* release-dists/pypi/rules/
           cp -R sdks/python/dist release-dists/sdks/python/
           cp -R server/dist release-dists/server/
 
-          for contrib_dir in evaluators/contrib/*/dist; do
+          for contrib_dir in rules/contrib/*/dist; do
             contrib_name="$(basename "$(dirname "$contrib_dir")")"
-            mkdir -p "release-dists/evaluators/contrib/$contrib_name"
-            cp -R "$contrib_dir" "release-dists/evaluators/contrib/$contrib_name/"
-            cp "$contrib_dir"/* release-dists/pypi/evaluators/
+            mkdir -p "release-dists/rules/contrib/$contrib_name"
+            cp -R "$contrib_dir" "release-dists/rules/contrib/$contrib_name/"
+            cp "$contrib_dir"/* release-dists/pypi/rules/
           done
 
       - name: Upload built distributions
@@ -101,7 +101,7 @@ jobs:
           user: __token__
           password: ${{ secrets.PYPI_API_TOKEN }}
 
-  publish-evaluators:
+  publish-rules:
     runs-on: ubuntu-latest
     needs: [release, publish-models]
     if: needs.release.outputs.released == 'true'
@@ -115,20 +115,20 @@ jobs:
           name: release-dists
           path: release-dists
 
-      - name: Publish evaluator distributions to PyPI
+      - name: Publish rule distributions to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1
         with:
-          packages-dir: release-dists/pypi/evaluators/
+          packages-dir: release-dists/pypi/rules/
           user: __token__
           password: ${{ secrets.PYPI_API_TOKEN }}
 
   publish-server:
     runs-on: ubuntu-latest
-    needs: [release, publish-evaluators]
+    needs: [release, publish-rules]
     if: >-
       always() &&
       needs.release.outputs.released == 'true' &&
-      needs.publish-evaluators.result == 'success'
+      needs.publish-rules.result == 'success'
     permissions:
       id-token: write
 
@@ -148,11 +148,11 @@ jobs:
 
   publish-sdk:
     runs-on: ubuntu-latest
-    needs: [release, publish-evaluators]
+    needs: [release, publish-rules]
     if: >-
       always() &&
       needs.release.outputs.released == 'true' &&
-      needs.publish-evaluators.result == 'success'
+      needs.publish-rules.result == 'success'
     permissions:
       id-token: write
 
@@ -172,7 +172,7 @@ jobs:
 
   upload-release-assets:
     runs-on: ubuntu-latest
-    needs: [release, publish-models, publish-evaluators, publish-server, publish-sdk]
+    needs: [release, publish-models, publish-rules, publish-server, publish-sdk]
     if: >-
       always() &&
       needs.release.outputs.released == 'true'
@@ -194,10 +194,10 @@ jobs:
           shopt -s nullglob
           assets=(
             release-dists/models/dist/*
-            release-dists/evaluators/builtin/dist/*
+            release-dists/rules/builtin/dist/*
             release-dists/sdks/python/dist/*
             release-dists/server/dist/*
-            release-dists/evaluators/contrib/*/dist/*
+            release-dists/rules/contrib/*/dist/*
           )
 
           if [ ${#assets[@]} -eq 0 ]; then

AGENTS.md

@@ -22,12 +22,12 @@ Forwarded targets:
 
 ## Repo layout (uv workspace members)
 
-- `models/`: shared Pydantic v2 models and evaluator base classes (`models/src/agent_control_models/`)
-- `engine/`: **control evaluation engine and evaluator system** — all evaluation logic, evaluator discovery, and evaluator orchestration lives here (`engine/src/agent_control_engine/`)
+- `models/`: shared Pydantic v2 models and rule base classes (`models/src/agent_control_models/`)
+- `engine/`: **control evaluation engine and rule system** — all evaluation logic, rule discovery, and rule orchestration lives here (`engine/src/agent_control_engine/`)
 - `server/`: FastAPI server (`server/src/agent_control_server/`)
 - `sdks/python/`: Python SDK — uses engine for evaluation (`sdks/python/src/agent_control/`)
-- `evaluators/builtin/`: builtin evaluator implementations (`evaluators/builtin/src/agent_control_evaluators/`)
-- `evaluators/contrib/`: optional evaluator packages (e.g., `evaluators/contrib/galileo/`)
+- `rules/builtin/`: builtin rule implementations (`rules/builtin/src/agent_control_rules/`)
+- `rules/contrib/`: optional rule packages (e.g., `rules/contrib/galileo/`)
 - `ui/`: Nextjs based web app to manage agent controls 
 - `examples/`: runnable examples (ruff has relaxed import rules here)
 
@@ -67,16 +67,16 @@ All testing guidance (including "behavior changes require tests") lives in `TEST
   4) add SDK wrapper in `sdks/python/src/agent_control/`
   5) add tests (server + SDK) and update docs/examples if user-facing
 
-- Add a new builtin evaluator:
-  1) implement evaluator class extending `Evaluator` in `evaluators/builtin/src/agent_control_evaluators/`
-  2) use `@register_evaluator` decorator (from `agent_control_evaluators`)
-  3) add entry point in `evaluators/builtin/pyproject.toml` for auto-discovery
-  4) add tests in the evaluators/builtin package
-  5) evaluator is automatically available to server and SDK via `discover_evaluators()`
+- Add a new builtin rule:
+  1) implement rule class extending `Rule` in `rules/builtin/src/agent_control_rules/`
+  2) use `@register_rule` decorator (from `agent_control_rules`)
+  3) add entry point in `rules/builtin/pyproject.toml` for auto-discovery
+  4) add tests in the rules/builtin package
+  5) rule is automatically available to server and SDK via `discover_rules()`
 
-- Add an external evaluator package:
-  1) copy `evaluators/contrib/template/` as a starting point
-  2) implement evaluator class extending `Evaluator` from `agent_control_evaluators`
+- Add an external rule package:
+  1) copy `rules/contrib/template/` as a starting point
+  2) implement rule class extending `Rule` from `agent_control_rules`
   3) add entry point using `org.name` format (e.g., `galileo.luna`)
   4) package is discovered automatically when installed alongside agent-control