fix: resolve 45 test failures across CLI, services, and assessors

src/agentready/cli/extract_skills.py

@@ -49,7 +49,7 @@
 )
 @click.option(
     "--llm-budget",
-    type=int,
+    type=click.IntRange(min=1),
     default=5,
     help="Maximum number of skills to enrich with LLM (default: 5)",
 )
@@ -153,10 +153,15 @@ def extract_skills(
             enable_llm = False
     click.echo()
 
+    # Resolve output directory relative to repository path if it's a relative path
+    output_dir_path = Path(output_dir)
+    if not output_dir_path.is_absolute():
+        output_dir_path = repo_path / output_dir
+
     # Create learning service
     learning_service = LearningService(
         min_confidence=min_confidence,
-        output_dir=output_dir,
+        output_dir=output_dir_path,
     )
 
     # Run learning workflow

src/agentready/cli/learn.py

@@ -49,7 +49,7 @@
 )
 @click.option(
     "--llm-budget",
-    type=int,
+    type=click.IntRange(min=1),
     default=5,
     help="Maximum number of skills to enrich with LLM (default: 5)",
 )
@@ -153,10 +153,15 @@ def learn(
             enable_llm = False
     click.echo()
 
+    # Resolve output directory relative to repository path if it's a relative path
+    output_dir_path = Path(output_dir)
+    if not output_dir_path.is_absolute():
+        output_dir_path = repo_path / output_dir
+
     # Create learning service
     learning_service = LearningService(
         min_confidence=min_confidence,
-        output_dir=output_dir,
+        output_dir=output_dir_path,
     )
 
     # Run learning workflow

src/agentready/learners/code_sampler.py

@@ -104,7 +104,7 @@ def _format_code_samples(self, files: list) -> str:
         samples = []
 
         for file_item in files[: self.max_files]:
-            if isinstance(file_item, dict):
+            if isinstance(file_item, dict) and "path" in file_item:
                 # Directory tree
                 samples.append(f"## Directory Structure: {file_item['path']}\n")
                 samples.append(self._format_tree(file_item))

src/agentready/learners/pattern_extractor.py

@@ -93,9 +93,8 @@ def extract_specific_patterns(
         discovered_skills = []
 
         for finding in self.assessment.findings:
-            if (
-                finding.attribute.attribute_id in attribute_ids
-                and self._should_extract_pattern(finding)
+            if finding.attribute.id in attribute_ids and self._should_extract_pattern(
+                finding
             ):
                 skill = self._create_skill_from_finding(finding)
                 if skill:

src/agentready/services/research_formatter.py

@@ -276,13 +276,12 @@ def format_report(self, content: str) -> str:
         lines = [line.rstrip() for line in lines]
         content = "\n".join(lines)
 
-        # Ensure file ends with single newline
-        if not content.endswith("\n"):
-            content += "\n"
-
         # Remove multiple blank lines (max 2 consecutive blank lines)
         content = re.sub(r"\n{4,}", "\n\n\n", content)
 
+        # Ensure file ends with exactly one newline
+        content = content.rstrip("\n") + "\n"
+
         return content
 
     def extract_attribute_ids(self, content: str) -> list[str]:
@@ -294,7 +293,8 @@ def extract_attribute_ids(self, content: str) -> list[str]:
         Returns:
             List of attribute IDs (e.g., ["1.1", "1.2", "2.1", ...])
         """
-        pattern = r"^###\s+(\d+\.\d+)\s+"
+        # Extract both valid and potentially malformed IDs for validation
+        pattern = r"^###\s+([\d]+\.[\w]+)\s+"
         matches = re.findall(pattern, content, re.MULTILINE)
         return matches
 

tests/unit/learners/test_llm_enricher.py

@@ -292,7 +292,12 @@ def test_enrich_skill_rate_limit_retry(
     client = Mock(spec=Anthropic)
 
     # First call raises rate limit, second succeeds
-    rate_limit_error = RateLimitError("Rate limit")
+    # Mock response and body for RateLimitError
+    mock_response = Mock()
+    mock_response.status_code = 429
+    rate_limit_error = RateLimitError(
+        "Rate limit", response=mock_response, body={"error": "rate_limit"}
+    )
     rate_limit_error.retry_after = 1  # 1 second retry
 
     success_response = Mock()
@@ -322,7 +327,12 @@ def test_enrich_skill_api_error_specific(
     from anthropic import APIError
 
     client = Mock(spec=Anthropic)
-    client.messages.create.side_effect = APIError("API Error")
+    # Mock request for APIError
+    mock_request = Mock()
+    mock_request.method = "POST"
+    client.messages.create.side_effect = APIError(
+        "API Error", request=mock_request, body={"error": "api_error"}
+    )
 
     cache_dir = tmp_path / "cache"
     enricher = LLMEnricher(client, cache_dir=cache_dir)

tests/unit/learners/test_pattern_extractor.py

@@ -9,6 +9,29 @@
 from agentready.models import Assessment, Attribute, Finding, Repository
 
 
+def create_dummy_finding() -> Finding:
+    """Create a dummy finding for testing (not_applicable status)."""
+    attr = Attribute(
+        id="test_attr",
+        name="Test Attribute",
+        category="Testing",
+        tier=1,
+        description="Test attribute",
+        criteria="Test criteria",
+        default_weight=1.0,
+    )
+    return Finding(
+        attribute=attr,
+        status="not_applicable",
+        score=None,
+        measured_value=None,
+        threshold=None,
+        evidence=[],
+        remediation=None,
+        error_message=None,
+    )
+
+
 def create_test_repository(tmp_path=None):
     """Create a test repository with valid path."""
     if tmp_path is None:
@@ -333,7 +356,7 @@ def test_should_extract_pattern_logic(self, sample_finding_high_score):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )
@@ -395,7 +418,7 @@ def test_create_skill_from_finding(self, sample_finding_high_score):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )
@@ -513,7 +536,7 @@ def test_extract_code_examples_from_evidence(self, sample_finding_high_score):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )
@@ -554,7 +577,7 @@ def test_extract_code_examples_limits_to_three(self, sample_repository):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )
@@ -574,7 +597,7 @@ def test_create_pattern_summary(self, sample_finding_high_score):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )
@@ -615,7 +638,7 @@ def test_pattern_summary_fallback_to_evidence(self, sample_repository):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=[create_dummy_finding()],
             config=None,
             duration_seconds=1.0,
         )

tests/unit/learners/test_skill_generator.py

@@ -128,7 +128,7 @@ def test_generate_markdown_report(self, sample_skill, tmp_path):
         content = output_file.read_text()
         assert "Test Skill" in content
         assert "test-skill" in content
-        assert "90%" in content  # Confidence
+        assert "90.0%" in content  # Confidence
         assert "+50.0 pts" in content  # Impact
         assert "85.0%" in content  # Reusability
 

tests/unit/test_cli_extract_skills.py

@@ -84,9 +84,9 @@ def test_extract_skills_command_skill_md_output(self, runner, temp_repo):
 
             assert result.exit_code == 0
 
-            # Check for SKILL.md files
+            # Check for SKILL.md files (in subdirectories: skill-id/SKILL.md)
             output_dir = temp_repo / ".skills-proposals"
-            md_files = list(output_dir.glob("*.md"))
+            md_files = list(output_dir.glob("*/SKILL.md"))
             assert len(md_files) > 0
 
     def test_extract_skills_command_github_issues_output(self, runner, temp_repo):
@@ -98,9 +98,9 @@ def test_extract_skills_command_github_issues_output(self, runner, temp_repo):
 
             assert result.exit_code == 0
 
-            # Check for issue files
+            # Check for issue files (named skill-{id}.md)
             output_dir = temp_repo / ".skills-proposals"
-            issue_files = list(output_dir.glob("issue-*.md"))
+            issue_files = list(output_dir.glob("skill-*.md"))
             assert len(issue_files) > 0
 
     def test_extract_skills_command_all_output_formats(self, runner, temp_repo):

tests/unit/test_cli_learn.py

@@ -25,13 +25,36 @@ def temp_repo():
         agentready_dir = repo_path / ".agentready"
         agentready_dir.mkdir()
 
-        # Create sample assessment using shared fixture
+        # Create sample assessment with known skill IDs that PatternExtractor recognizes
+        from tests.fixtures.assessment_fixtures import create_test_finding_json
+
+        findings = [
+            create_test_finding_json(
+                attribute_id="claude_md_file",
+                attribute_name="CLAUDE.md File",
+                status="pass",
+                score=95.0,
+                category="Documentation",
+                tier=1,
+            ),
+            create_test_finding_json(
+                attribute_id="type_annotations",
+                attribute_name="Type Annotations",
+                status="pass",
+                score=90.0,
+                category="Code Quality",
+                tier=2,
+            ),
+        ]
+
         assessment_data = create_test_assessment_json(
             overall_score=85.0,
             num_findings=2,
             repo_path=str(repo_path),
             repo_name="test-repo",
         )
+        # Replace generic findings with skill-specific ones
+        assessment_data["findings"] = findings
 
         assessment_file = agentready_dir / "assessment-latest.json"
         with open(assessment_file, "w") as f:
@@ -82,9 +105,9 @@ def test_learn_command_skill_md_output(self, runner, temp_repo):
 
             assert result.exit_code == 0
 
-            # Check for SKILL.md files
+            # Check for SKILL.md files (in subdirectories: skill-id/SKILL.md)
             output_dir = temp_repo / ".skills-proposals"
-            md_files = list(output_dir.glob("*.md"))
+            md_files = list(output_dir.glob("*/SKILL.md"))
             assert len(md_files) > 0
 
     def test_learn_command_github_issues_output(self, runner, temp_repo):
@@ -96,9 +119,9 @@ def test_learn_command_github_issues_output(self, runner, temp_repo):
 
             assert result.exit_code == 0
 
-            # Check for issue files
+            # Check for issue files (named skill-{id}.md)
             output_dir = temp_repo / ".skills-proposals"
-            issue_files = list(output_dir.glob("issue-*.md"))
+            issue_files = list(output_dir.glob("skill-*.md"))
             assert len(issue_files) > 0
 
     def test_learn_command_all_output_formats(self, runner, temp_repo):

tests/unit/test_csv_reporter.py

@@ -6,15 +6,44 @@
 import pytest
 
 from src.agentready.models.assessment import Assessment
+from src.agentready.models.attribute import Attribute
 from src.agentready.models.batch_assessment import (
     BatchAssessment,
     BatchSummary,
     RepositoryResult,
 )
+from src.agentready.models.finding import Finding
 from src.agentready.models.repository import Repository
 from src.agentready.reporters.csv_reporter import CSVReporter
 
 
+def create_dummy_findings(count: int) -> list[Finding]:
+    """Create dummy findings for testing."""
+    findings = []
+    for i in range(count):
+        attr = Attribute(
+            id=f"test_attr_{i}",
+            name=f"Test Attribute {i}",
+            category="Testing",
+            tier=1,
+            description="Test attribute",
+            criteria="Test criteria",
+            default_weight=1.0,
+        )
+        finding = Finding(
+            attribute=attr,
+            status="not_applicable",
+            score=None,
+            measured_value=None,
+            threshold=None,
+            evidence=[],
+            remediation=None,
+            error_message=None,
+        )
+        findings.append(finding)
+    return findings
+
+
 @pytest.fixture
 def temp_csv_file(tmp_path):
     """Create temporary CSV file for testing."""
@@ -104,7 +133,7 @@ def mock_batch_assessment(mock_assessment, tmp_path):
         attributes_assessed=20,
         attributes_not_assessed=5,
         attributes_total=25,
-        findings=[],
+        findings=create_dummy_findings(25),
         config=None,
         duration_seconds=38.0,
         discovered_skills=[],
@@ -293,20 +322,16 @@ def test_csv_empty_batch(self, tmp_path):
             failed_assessments=0,
             average_score=0.0,
         )
-        batch = BatchAssessment(
-            batch_id="empty-batch",
-            timestamp=datetime.now(),
-            results=[],
-            summary=summary,
-            total_duration_seconds=0.0,
-        )
-
-        reporter = CSVReporter()
-        output_path = tmp_path / "empty.csv"
 
-        # Should not raise error, but will only have header
+        # BatchAssessment validation should raise ValueError during construction
         with pytest.raises(ValueError, match="Batch must have at least one result"):
-            reporter.generate(batch, output_path)
+            BatchAssessment(
+                batch_id="empty-batch",
+                timestamp=datetime.now(),
+                results=[],
+                summary=summary,
+                total_duration_seconds=0.0,
+            )
 
     def test_csv_creates_parent_directory(self, tmp_path):
         """Test that CSV reporter creates parent directories if needed."""
@@ -336,7 +361,7 @@ def test_csv_creates_parent_directory(self, tmp_path):
             attributes_assessed=1,
             attributes_not_assessed=0,
             attributes_total=1,
-            findings=[],
+            findings=create_dummy_findings(1),
             config=None,
             duration_seconds=1.0,
             discovered_skills=[],