drydock-cli 1.3.8__tar.gz → 1.4.0__tar.gz

{drydock_cli-1.3.8 → drydock_cli-1.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: drydock-cli
-Version: 1.3.8
+Version: 1.4.0
 Summary: Drydock — nautical CLI coding agent
 Project-URL: Homepage, https://github.com/fbobe321/drydock
 Project-URL: Repository, https://github.com/fbobe321/drydock

{drydock_cli-1.3.8 → drydock_cli-1.4.0}/drydock/core/agent_loop.py

@@ -617,25 +617,22 @@ class AgentLoop:
                     )
 
                 # Bash abuse detection: model uses bash instead of proper tools
-                # This is the #1 cause of no-patch failures (88% of cases)
-                if not has_made_edit and bash_count >= 5:
-                    if bash_count == 5:
+                if not has_made_edit and bash_count >= 3:
+                    if bash_count == 3:
                         self._inject_system_note(
-                            "You have run 5 bash commands without making any edit. "
-                            "STOP using bash for reading/searching. Use the proper tools: "
-                            "read_file (not cat), grep (not bash grep), search_replace (not sed). "
-                            "Call search_replace NOW to fix the bug."
+                            "You have run 3 bash commands without creating or editing any files. "
+                            "STOP running bash. Start CREATING files with write_file or "
+                            "EDITING files with search_replace. bash does not create code."
                         )
-                    elif bash_count == 8:
+                    elif bash_count == 6:
                         self._inject_system_note(
-                            "WARNING: 8 bash commands, ZERO edits. You are wasting context. "
-                            "Bash does NOT fix bugs — search_replace does. "
-                            "Use search_replace on your NEXT response or the session ends."
+                            "WARNING: 6 bash commands, ZERO file edits. You are stuck in a loop. "
+                            "Your NEXT action MUST be write_file to create a file or "
+                            "search_replace to edit one. Do NOT run any more bash commands."
                         )
-                    elif bash_count >= 12:
-                        # Force stop — model is clearly stuck in bash loop
+                    elif bash_count >= 10:
                         yield AssistantEvent(
-                            content="\n\n[STOPPED: 12+ bash commands without any edit. Use search_replace.]\n",
+                            content="\n\n[STOPPED: 10+ bash commands without creating any files. Use write_file.]\n",
                             stopped_by_middleware=True,
                         )
                         return

{drydock_cli-1.3.8 → drydock_cli-1.4.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "drydock-cli"
-version = "1.3.8"
+version = "1.4.0"
 description = "Drydock — nautical CLI coding agent"
 readme = "README.md"
 requires-python = ">=3.12"

{drydock_cli-1.3.8 → drydock_cli-1.4.0}/scripts/test_full.sh

@@ -11,3 +11,7 @@ python3 -m pytest tests/test_smoke.py -p no:xdist -p no:cov --override-ini="addo
 echo ""
 echo "=== FULL REGRESSION (real backend) ==="
 python3 -m pytest tests/test_full_regression.py -v -p no:xdist -p no:cov --override-ini="addopts="
+
+echo ""
+echo "=== BUILD PROJECT TESTS (real backend, 30-60 min) ==="
+python3 -m pytest tests/test_build_projects.py -v -s -p no:xdist -p no:cov --override-ini="addopts="

drydock_cli-1.4.0/tests/test_build_projects.py

@@ -0,0 +1,272 @@
+"""Build project tests — DryDock must actually BUILD things.
+
+These are LONG tests (5-20 min each) that run against the real vLLM backend.
+They ask DryDock to build real projects and verify:
+1. Files are actually created
+2. No circuit breaker kills the session
+3. No message ordering crashes
+4. No infinite loops (bash abuse, repetition)
+5. Created code is valid Python (syntax check)
+
+Run: pytest tests/test_build_projects.py -v -s
+Expected runtime: 30-60 minutes total
+"""
+
+from __future__ import annotations
+
+import asyncio
+import os
+from pathlib import Path
+
+import httpx
+import pytest
+
+from drydock.core.config.harness_files import init_harness_files_manager
+try:
+    init_harness_files_manager("user", "project")
+except RuntimeError:
+    pass
+
+from drydock.core.agent_loop import AgentLoop
+from drydock.core.agents.models import BuiltinAgentName
+from drydock.core.config import Backend, ModelConfig, ProviderConfig, VibeConfig
+from drydock.core.types import AssistantEvent, ToolCallEvent, ToolResultEvent
+
+
+def _vllm_ok():
+    try:
+        return httpx.get("http://localhost:8000/v1/models", timeout=3).status_code == 200
+    except Exception:
+        return False
+
+pytestmark = pytest.mark.skipif(not _vllm_ok(), reason="vLLM not running")
+
+
+def _agent(work_dir: Path, max_turns: int = 30):
+    config = VibeConfig(
+        active_model="devstral", auto_approve=True, enable_telemetry=False,
+        include_project_context=False, system_prompt_id="cli",
+        providers=[ProviderConfig(name="local", api_base="http://localhost:8000/v1", api_key_env_var="", backend=Backend.GENERIC)],
+        models=[ModelConfig(name="devstral", provider="local", input_price=0, output_price=0)],
+        session_logging={"enabled": False, "save_dir": str(work_dir / ".logs")},
+    )
+    os.chdir(work_dir)
+    return AgentLoop(config=config, agent_name=BuiltinAgentName.AUTO_APPROVE, max_turns=max_turns)
+
+
+async def _run_and_evaluate(agent, prompt, work_dir, max_events=200):
+    """Run agent and collect detailed metrics."""
+    events = []
+    tool_counts = {}
+    errors = []
+    circuit_breaker_fires = 0
+    force_stops = 0
+    ordering_crashes = 0
+
+    async for ev in agent.act(prompt):
+        events.append(ev)
+        if isinstance(ev, ToolCallEvent):
+            tool_counts[ev.tool_name] = tool_counts.get(ev.tool_name, 0) + 1
+        elif isinstance(ev, ToolResultEvent) and ev.error:
+            err = str(ev.error)
+            errors.append(err[:100])
+            if "CIRCUIT BREAKER" in err:
+                circuit_breaker_fires += 1
+            if "Unexpected role" in err:
+                ordering_crashes += 1
+        elif isinstance(ev, AssistantEvent) and ev.stopped_by_middleware:
+            if "FORCED STOP" in (ev.content or ""):
+                force_stops += 1
+        if len(events) >= max_events:
+            break
+
+    # Count created files
+    py_files = list(work_dir.rglob("*.py"))
+    all_files = [f for f in work_dir.rglob("*") if f.is_file() and ".logs" not in str(f)]
+
+    # Syntax check created Python files
+    syntax_errors = 0
+    for f in py_files:
+        try:
+            import ast
+            ast.parse(f.read_text())
+        except SyntaxError:
+            syntax_errors += 1
+
+    return {
+        "events": len(events),
+        "tool_counts": tool_counts,
+        "errors": len(errors),
+        "circuit_breaker_fires": circuit_breaker_fires,
+        "force_stops": force_stops,
+        "ordering_crashes": ordering_crashes,
+        "py_files": len(py_files),
+        "all_files": len(all_files),
+        "syntax_errors": syntax_errors,
+        "error_details": errors[:5],
+    }
+
+
+# ============================================================================
+# Test 1: Build a CLI tool from a PRD
+# ============================================================================
+
+@pytest.mark.asyncio
+async def test_build_cli_from_prd(tmp_path):
+    """User scenario: 'review the PRD and get started'
+
+    Must create a Python package with multiple files.
+    Must NOT crash, loop, or get killed by circuit breaker.
+    """
+    (tmp_path / "PRD.md").write_text(
+        "# File Counter CLI\n\n"
+        "Build a Python CLI tool that counts files in a directory.\n\n"
+        "## Features\n"
+        "- Count files by extension\n"
+        "- Show total size\n"
+        "- Support --recursive flag\n\n"
+        "## Structure\n"
+        "- file_counter/ package\n"
+        "- counter.py — core counting logic\n"
+        "- cli.py — argparse interface\n"
+        "- __init__.py\n"
+    )
+
+    agent = _agent(tmp_path, max_turns=25)
+    result = await _run_and_evaluate(agent, "review the PRD and build the project", tmp_path)
+
+    print(f"\n=== BUILD CLI RESULTS ===")
+    print(f"Events: {result['events']}, Tools: {result['tool_counts']}")
+    print(f"Files: {result['py_files']} Python, {result['all_files']} total")
+    print(f"Errors: {result['errors']}, CB: {result['circuit_breaker_fires']}, Stops: {result['force_stops']}")
+    if result['error_details']:
+        print(f"Error samples: {result['error_details'][:3]}")
+
+    assert result["ordering_crashes"] == 0, "Message ordering crash"
+    assert result["force_stops"] == 0, "Session force-stopped"
+    assert result["circuit_breaker_fires"] <= 2, f"Circuit breaker fired {result['circuit_breaker_fires']} times"
+    assert result["py_files"] >= 2, f"Only {result['py_files']} Python files created (need 2+)"
+    assert result["syntax_errors"] == 0, f"{result['syntax_errors']} files have syntax errors"
+
+
+# ============================================================================
+# Test 2: Build a simple Python script
+# ============================================================================
+
+@pytest.mark.asyncio
+async def test_build_simple_script(tmp_path):
+    """User scenario: 'write a Python script that...'
+
+    Simpler task — should complete quickly with write_file.
+    """
+    agent = _agent(tmp_path, max_turns=15)
+    result = await _run_and_evaluate(
+        agent,
+        "Write a Python script called calculator.py that implements add, subtract, multiply, divide functions and a main() that demos them",
+        tmp_path,
+    )
+
+    print(f"\n=== SIMPLE SCRIPT RESULTS ===")
+    print(f"Events: {result['events']}, Tools: {result['tool_counts']}")
+    print(f"Files: {result['py_files']} Python")
+
+    assert result["ordering_crashes"] == 0
+    assert result["force_stops"] == 0
+    assert result["py_files"] >= 1, "No Python file created"
+    assert (tmp_path / "calculator.py").exists(), "calculator.py not created"
+
+
+# ============================================================================
+# Test 3: Add a feature to existing code
+# ============================================================================
+
+@pytest.mark.asyncio
+async def test_add_feature_to_existing(tmp_path):
+    """User scenario: 'add X to this code'
+
+    Tests search_replace on existing files.
+    """
+    (tmp_path / "app.py").write_text(
+        "def greet(name):\n"
+        "    return f'Hello, {name}!'\n\n"
+        "if __name__ == '__main__':\n"
+        "    print(greet('World'))\n"
+    )
+
+    agent = _agent(tmp_path, max_turns=15)
+    result = await _run_and_evaluate(
+        agent,
+        "Add a goodbye function to app.py and call it from main",
+        tmp_path,
+    )
+
+    print(f"\n=== ADD FEATURE RESULTS ===")
+    print(f"Events: {result['events']}, Tools: {result['tool_counts']}")
+
+    assert result["ordering_crashes"] == 0
+    assert result["force_stops"] == 0
+    # Should have used search_replace or write_file
+    edit_tools = result["tool_counts"].get("search_replace", 0) + result["tool_counts"].get("write_file", 0)
+    assert edit_tools >= 1, f"No edits made (search_replace={result['tool_counts'].get('search_replace',0)}, write_file={result['tool_counts'].get('write_file',0)})"
+
+    content = (tmp_path / "app.py").read_text()
+    assert "goodbye" in content.lower() or "bye" in content.lower(), "Feature not added"
+
+
+# ============================================================================
+# Test 4: Fix a bug in existing code
+# ============================================================================
+
+@pytest.mark.asyncio
+async def test_fix_bug(tmp_path):
+    """User scenario: 'there's a bug, fix it'"""
+    (tmp_path / "divide.py").write_text(
+        "def safe_divide(a, b):\n"
+        "    return a / b  # BUG: no zero check\n\n"
+        "print(safe_divide(10, 0))\n"
+    )
+
+    agent = _agent(tmp_path, max_turns=15)
+    result = await _run_and_evaluate(
+        agent,
+        "Fix the divide by zero bug in divide.py",
+        tmp_path,
+    )
+
+    print(f"\n=== FIX BUG RESULTS ===")
+    print(f"Events: {result['events']}, Tools: {result['tool_counts']}")
+
+    assert result["ordering_crashes"] == 0
+    assert result["force_stops"] == 0
+
+    content = (tmp_path / "divide.py").read_text()
+    has_fix = "if" in content or "try" in content or "== 0" in content or "ZeroDivision" in content
+    assert has_fix, "Bug not fixed — no zero check added"
+
+
+# ============================================================================
+# Test 5: Multi-file project with tests
+# ============================================================================
+
+@pytest.mark.asyncio
+async def test_build_project_with_tests(tmp_path):
+    """User scenario: 'build a project with tests'
+
+    The hardest test — must create package + test files.
+    """
+    agent = _agent(tmp_path, max_turns=30)
+    result = await _run_and_evaluate(
+        agent,
+        "Create a Python package called 'mathlib' with functions for factorial, fibonacci, and is_prime. Include a test file.",
+        tmp_path,
+    )
+
+    print(f"\n=== PROJECT WITH TESTS RESULTS ===")
+    print(f"Events: {result['events']}, Tools: {result['tool_counts']}")
+    print(f"Files: {result['py_files']} Python, {result['all_files']} total")
+    print(f"Syntax errors: {result['syntax_errors']}")
+
+    assert result["ordering_crashes"] == 0
+    assert result["force_stops"] == 0
+    assert result["py_files"] >= 2, f"Only {result['py_files']} Python files (need package + tests)"
+    assert result["syntax_errors"] == 0, "Created files have syntax errors"