Restructure

.github/workflows/fix-remote-pr.yml

@@ -244,7 +244,7 @@ jobs:
           cd ../bot-repo
 
           echo "Running classifier..."
-          if ! classify-pr-failure \
+          if ! aieng-bot classify \
             --pr-info "$PR_INFO" \
             --failed-checks "$FAILED_CHECKS" \
             --failure-logs-file /tmp/failure-logs.txt \
@@ -360,7 +360,7 @@ jobs:
         working-directory: target-repo
         run: |
           # Apply fixes using skills from .claude/skills/
-          apply-agent-fix \
+          aieng-bot fix \
             --repo "${{ github.event.inputs.target_repo }}" \
             --pr-number "${{ github.event.inputs.pr_number }}" \
             --pr-title "${{ steps.pr-details.outputs.pr-title }}" \

.github/workflows/monitor-org-bot-prs.yml

@@ -135,7 +135,7 @@ jobs:
 
       - name: Process repository queue
         run: |
-          process-repo-queue \
+          aieng-bot queue \
             --repo "${{ matrix.repo }}" \
             --workflow-run-id "${{ github.run_id }}" \
             --all-prs '${{ needs.discover-bot-prs.outputs.prs }}'

.github/workflows/unit_tests.yml

@@ -47,7 +47,7 @@ jobs:
 
       - name: Run unit tests with coverage
         run: |
-          uv run pytest --cov src/aieng_bot_maintain --cov-report=xml --cov-report=term tests/
+          uv run pytest --cov src/aieng_bot --cov-report=xml --cov-report=term tests/
 
       - name: Upload coverage to Codecov
         uses: codecov/codecov-action@v5

.gitignore

@@ -48,5 +48,5 @@ env/
 # Project-specific
 CLAUDE.md
 
-aieng_bot_maintain.egg-info/
+aieng_bot.egg-info/
 coverage.xml

README.md

@@ -170,7 +170,7 @@ gh run list --workflow=monitor-org-bot-prs.yml --limit 5
 gh run view RUN_ID --log
 
 # Collect metrics manually
-gh workflow run collect-bot-metrics.yml
+gh workflow run aieng-bot metrics.yml
 ```
 
 ## Documentation

docs/index.md

@@ -170,7 +170,7 @@ gh run list --workflow=monitor-org-bot-prs.yml --limit 5
 gh run view RUN_ID --log
 
 # Collect metrics manually
-gh workflow run collect-bot-metrics.yml
+gh workflow run aieng-bot metrics.yml
 ```
 
 ## Documentation

pyproject.toml

@@ -3,7 +3,7 @@ requires = ["hatchling"]
 build-backend = "hatchling.build"
 
 [project]
-name = "aieng-bot-maintain"
+name = "aieng-bot"
 version = "0.4.0"
 description = "Vector Institute AI Engineering Bot for Maintenance Tasks"
 readme = "README.md"
@@ -23,6 +23,7 @@ classifiers = [
 dependencies = [
     "anthropic>=0.42.0",
     "claude-agent-sdk",
+    "click>=8.0.0",
     "pyyaml>=6.0.2",
     "rich>=13.0.0",
 ]
@@ -34,16 +35,13 @@ Issues = "https://github.com/VectorInstitute/aieng-bot-maintain/issues"
 Changelog = "https://github.com/VectorInstitute/aieng-bot-maintain/releases"
 
 [project.scripts]
-classify-pr-failure = "aieng_bot_maintain.cli:classify_pr_failure_cli"
-apply-agent-fix = "aieng_bot_maintain.cli:apply_agent_fix_cli"
-collect-bot-metrics = "aieng_bot_maintain.cli:collect_metrics_cli"
-process-repo-queue = "aieng_bot_maintain.cli:process_repo_queue_cli"
+aieng-bot = "aieng_bot._cli.main:cli"
 
 [tool.hatchling.build.targets.wheel]
-packages = ["src/aieng_bot_maintain"]
+packages = ["src/aieng_bot"]
 
 [tool.hatchling.build.targets.wheel.shared-data]
-"src/aieng_bot_maintain/py.typed" = "aieng_bot_maintain/py.typed"
+"src/aieng_bot/py.typed" = "aieng_bot/py.typed"
 
 [dependency-groups]
 dev = [

slack_bot/app.py

@@ -17,10 +17,10 @@
 # Load environment variables from .env file
 load_dotenv()
 
-# Add parent directory to path to import from aieng_bot_maintain
+# Add parent directory to path to import from aieng_bot
 sys.path.insert(0, str(Path(__file__).parent.parent / "src"))
 
-from aieng_bot_maintain.utils.logging import (  # noqa: E402
+from aieng_bot.utils.logging import (  # noqa: E402
     log_error,
     log_info,
     log_success,

src/aieng_bot_maintain/__init__.py → src/aieng_bot/__init__.py

@@ -16,6 +16,7 @@
     FailureType,
     PRContext,
 )
+from .config import get_model_name
 from .metrics import MetricsCollector
 from .observability import AgentExecutionTracer, create_tracer_from_env
 
@@ -33,5 +34,6 @@
     "RepoQueue",
     "PRQueueItem",
     "PRStatus",
+    "get_model_name",
     "__version__",
 ]

src/aieng_bot/_cli/__init__.py

@@ -0,0 +1,5 @@
+"""CLI module for aieng-bot."""
+
+from .main import cli
+
+__all__ = ["cli"]

src/aieng_bot/_cli/commands/__init__.py

@@ -0,0 +1,8 @@
+"""CLI commands for aieng-bot."""
+
+from .classify import classify
+from .fix import fix
+from .metrics import metrics
+from .queue import queue
+
+__all__ = ["classify", "fix", "metrics", "queue"]

src/aieng_bot_maintain/_cli/commands/classify.py → src/aieng_bot/_cli/commands/classify.py

@@ -1,27 +1,30 @@
 """CLI command for PR failure classification."""
 
-import argparse
 import json
 import sys
 import tempfile
 
+import click
+
 from ...classifier import PRFailureClassifier
 from ...classifier.models import ClassificationResult
 from ...utils.logging import get_console, log_error, log_info, log_success
-from ..utils import get_version, parse_pr_inputs
+from ..utils import parse_pr_inputs
 
 
-def _get_failure_logs_file(args: argparse.Namespace) -> str | None:
-    """Get failure logs file path from args, return None on error."""
-    if args.failure_logs_file:
-        return args.failure_logs_file
+def _get_failure_logs_file(
+    failure_logs: str | None, failure_logs_file: str | None
+) -> str | None:
+    """Get failure logs file path, return None on error."""
+    if failure_logs_file:
+        return failure_logs_file
 
-    if args.failure_logs:
-        # Backward compatibility: write logs to temp file
+    if failure_logs:
+        # Write logs to temp file
         with tempfile.NamedTemporaryFile(
             mode="w", delete=False, suffix=".txt"
         ) as temp_file:
-            temp_file.write(args.failure_logs)
+            temp_file.write(failure_logs)
             return temp_file.name
 
     log_error("Either --failure-logs or --failure-logs-file must be provided")
@@ -61,80 +64,87 @@ def _log_summary(result: ClassificationResult) -> None:
         sys.exit(1)
 
 
-def classify_pr_failure_cli() -> None:
-    """CLI entry point for PR failure classification.
+@click.command()
+@click.option(
+    "--pr-info",
+    required=True,
+    help="PR info JSON string containing repo, pr_number, title, author, etc.",
+)
+@click.option(
+    "--failed-checks",
+    required=True,
+    help="Failed checks JSON array with check names and conclusions",
+)
+@click.option(
+    "--failure-logs",
+    required=False,
+    help="Failure logs content (truncated). Use --failure-logs-file for large logs.",
+)
+@click.option(
+    "--failure-logs-file",
+    required=False,
+    type=click.Path(exists=True),
+    help="Path to file containing failure logs (alternative to --failure-logs)",
+)
+@click.option(
+    "--output-format",
+    type=click.Choice(["json", "github"], case_sensitive=False),
+    default="github",
+    help="Output format: 'github' for GitHub Actions variables, 'json' for structured output",
+)
+def classify(
+    pr_info: str,
+    failed_checks: str,
+    failure_logs: str | None,
+    failure_logs_file: str | None,
+    output_format: str,
+) -> None:
+    r"""Classify PR failure type using Claude API.
+
+    Analyzes PR context, failed checks, and logs to determine failure category
+    (test, lint, security, build, merge_conflict, unknown).
 
-    Reads PR context, failed checks, and failure logs from command-line arguments
-    and outputs classification results in GitHub Actions format or JSON.
+    Examples:
+      \b
+      # Classify with GitHub Actions output
+      aieng-bot classify --pr-info '$PR_JSON' --failed-checks '$CHECKS_JSON' \\
+        --failure-logs-file logs.txt
+
+      \b
+      # Classify with JSON output
+      aieng-bot classify --pr-info '$PR_JSON' --failed-checks '$CHECKS_JSON' \\
+        --failure-logs "$(cat logs.txt)" --output-format json
 
     """
-    parser = argparse.ArgumentParser(
-        prog="classify-pr-failure",
-        description="Classify PR failure type using Claude API",
-        formatter_class=argparse.RawDescriptionHelpFormatter,
-        epilog="""
-Examples:
-  # Classify with GitHub Actions output
-  classify-pr-failure --pr-info '$PR_JSON' --failed-checks '$CHECKS_JSON' \\
-    --failure-logs "$(cat logs.txt)"
-
-  # Classify with JSON output
-  classify-pr-failure --pr-info '$PR_JSON' --failed-checks '$CHECKS_JSON' \\
-    --failure-logs "$(cat logs.txt)" --output-format json
-        """,
-    )
-    parser.add_argument(
-        "--version",
-        action="version",
-        version=f"%(prog)s {get_version()}",
-        help="Show version number and exit",
-    )
-    parser.add_argument("--pr-info", required=True, help="PR info JSON string")
-    parser.add_argument(
-        "--failed-checks", required=True, help="Failed checks JSON array"
-    )
-    parser.add_argument(
-        "--failure-logs", required=False, help="Failure logs (truncated)"
-    )
-    parser.add_argument(
-        "--failure-logs-file",
-        required=False,
-        help="Path to file containing failure logs (alternative to --failure-logs)",
-    )
-    parser.add_argument(
-        "--output-format",
-        choices=["json", "github"],
-        default="github",
-        help="Output format (default: github)",
-    )
-
-    args = parser.parse_args()
     console = get_console()
 
     try:
-        # Parse inputs
-        pr_context, failed_checks = parse_pr_inputs(args)
+        # Parse inputs - create argparse.Namespace for compatibility with parse_pr_inputs
+        import argparse  # noqa: PLC0415
+
+        args = argparse.Namespace(pr_info=pr_info, failed_checks=failed_checks)
+        pr_context, failed_check_list = parse_pr_inputs(args)
 
         # Get failure logs file path
-        failure_logs_file = _get_failure_logs_file(args)
-        if not failure_logs_file:
+        failure_logs_path = _get_failure_logs_file(failure_logs, failure_logs_file)
+        if not failure_logs_path:
             sys.exit(1)
 
         # Run classification
         log_info(f"Classifying PR {pr_context.repo}#{pr_context.pr_number}")
-        log_info(f"Number of failed checks: {len(failed_checks)}")
-        log_info(f"Failure logs file: {failure_logs_file}")
+        log_info(f"Number of failed checks: {len(failed_check_list)}")
+        log_info(f"Failure logs file: {failure_logs_path}")
 
         classifier = PRFailureClassifier()
-        result = classifier.classify(pr_context, failed_checks, failure_logs_file)
+        result = classifier.classify(pr_context, failed_check_list, failure_logs_path)
 
         log_info(
             f"Classification result: {result.failure_type.value} "
             f"(confidence: {result.confidence:.2f})"
         )
 
         # Output results
-        _output_results(result, args.output_format, console)
+        _output_results(result, output_format, console)
 
         # Log summary
         _log_summary(result)