codespy-ai 0.2.0__tar.gz → 0.2.2__tar.gz

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codespy-ai
-Version: 0.2.0
+Version: 0.2.2
 Summary: Code review agent powered by DSPy
 License: MIT
 License-File: LICENSE
@@ -247,13 +247,44 @@ docker run --rm \
   -e GITHUB_TOKEN=$GITHUB_TOKEN \
   -e DEFAULT_MODEL=claude-opus-4-6 \
   -e ANTHROPIC_API_KEY=$ANTHROPIC_API_KEY \
-  ghcr.io/khezen/codespy:0.1.0 review https://github.com/owner/repo/pull/123
+  ghcr.io/khezen/codespy:0.2.1 review https://github.com/owner/repo/pull/123
 ```
 
 ### GitHub Action
 
 Add CodeSpy to your repository for automatic PR reviews:
 
+**Trigger on `@codespy review` comment:**
+
+```yaml
+# .github/workflows/codespy-review.yml
+name: CodeSpy Code Review
+
+on:
+  issue_comment:
+    types: [created]
+
+jobs:
+  review:
+    # Only run on PR comments containing '@codespy review'
+    if: |
+      github.event.issue.pull_request &&
+      contains(github.event.comment.body, '@codespy review')
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: write
+
+    steps:
+      - name: Run CodeSpy Review
+        uses: khezen/codespy@v1
+        with:
+          model: 'claude-opus-4-6'
+          anthropic-api-key: ${{ secrets.ANTHROPIC_API_KEY }}
+```
+
+**Trigger automatically on every PR:**
+
 ```yaml
 # .github/workflows/codespy-review.yml
 name: CodeSpy Code Review

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/README.md

@@ -196,13 +196,44 @@ docker run --rm \
   -e GITHUB_TOKEN=$GITHUB_TOKEN \
   -e DEFAULT_MODEL=claude-opus-4-6 \
   -e ANTHROPIC_API_KEY=$ANTHROPIC_API_KEY \
-  ghcr.io/khezen/codespy:0.1.0 review https://github.com/owner/repo/pull/123
+  ghcr.io/khezen/codespy:0.2.1 review https://github.com/owner/repo/pull/123
 ```
 
 ### GitHub Action
 
 Add CodeSpy to your repository for automatic PR reviews:
 
+**Trigger on `@codespy review` comment:**
+
+```yaml
+# .github/workflows/codespy-review.yml
+name: CodeSpy Code Review
+
+on:
+  issue_comment:
+    types: [created]
+
+jobs:
+  review:
+    # Only run on PR comments containing '@codespy review'
+    if: |
+      github.event.issue.pull_request &&
+      contains(github.event.comment.body, '@codespy review')
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: write
+
+    steps:
+      - name: Run CodeSpy Review
+        uses: khezen/codespy@v1
+        with:
+          model: 'claude-opus-4-6'
+          anthropic-api-key: ${{ secrets.ANTHROPIC_API_KEY }}
+```
+
+**Trigger automatically on every PR:**
+
 ```yaml
 # .github/workflows/codespy-review.yml
 name: CodeSpy Code Review

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "codespy-ai"
-version = "0.2.0"
+version = "0.2.2"
 description = "Code review agent powered by DSPy"
 readme = "README.md"
 license = "MIT"

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/src/codespy/__init__.py

@@ -1,3 +1,3 @@
 """codespy - Code review agent powered by DSPy."""
 
-__version__ = "0.1.0"
+__version__ = "0.2.2"

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/src/codespy/agents/reviewer/reviewer.py

@@ -1,5 +1,6 @@
 """Main review pipeline that orchestrates all review modules."""
 
+import asyncio
 import logging
 from pathlib import Path
 
@@ -100,6 +101,40 @@ class ReviewPipeline(dspy.Module):
         owner_path = mr.repo_owner.replace("/", "_")
         return cache_dir / owner_path / mr.repo_name
 
+    async def _run_review_modules(
+        self,
+        scopes: list,
+        repo_path: Path,
+        module_names: list[str],
+    ) -> list[Issue]:
+        """Run review modules concurrently in a single event loop.
+
+        Uses asyncio.gather instead of dspy.Parallel to avoid the
+        multi-thread + multi-event-loop conflict that causes
+        'cannot schedule new futures after shutdown' errors.
+
+        Args:
+            scopes: Identified scopes with changed files
+            repo_path: Path to the cloned repository
+            module_names: Names of modules (for error logging)
+
+        Returns:
+            Aggregated list of issues from all modules
+        """
+        tasks = [
+            self.code_and_doc_reviewer.aforward(scopes=scopes, repo_path=repo_path),
+            self.supply_chain_auditor.aforward(scopes=scopes, repo_path=repo_path),
+        ]
+        results = await asyncio.gather(*tasks, return_exceptions=True)
+
+        all_issues: list[Issue] = []
+        for i, result in enumerate(results):
+            if isinstance(result, Exception):
+                logger.error(f"{module_names[i]} failed: {result}")
+            elif result is not None:
+                all_issues.extend(result)
+        return all_issues
+
     def forward(self, mr_url: str, verify_model: bool = True) -> ReviewResult:
         """Run the complete review pipeline on a merge request."""
         self.cost_tracker.reset()
@@ -122,28 +157,13 @@ class ReviewPipeline(dspy.Module):
                 if manifest.dependencies_changed:
                     logger.info(f"    Dependencies changed: Yes")
 
-        # Build list of review modules (they check signature enabled status internally)
-        all_issues: list[Issue] = []
-        exec_pairs = [
-            (self.code_and_doc_reviewer, {"scopes": scopes, "repo_path": repo_path}),
-            (self.supply_chain_auditor, {"scopes": scopes, "repo_path": repo_path}),
-        ]
+        # Run review modules concurrently via asyncio.gather
         module_names = ["code_and_doc_reviewer", "supply_chain_auditor"]
-
-        logger.info(f"Running review modules in parallel: {', '.join(module_names)}...")
-        # Execute in parallel with error handling
-        if exec_pairs:
-            parallel = dspy.Parallel(num_threads=len(exec_pairs), return_failed_examples=True, provide_traceback=True)
-            results, failed_examples, exceptions = parallel(exec_pairs)
-            # Log any failures
-            for i, (failed, exc) in enumerate(zip(failed_examples, exceptions)):
-                if failed is not None and exc is not None:
-                    logger.error(f"{module_names[i]} failed: {exc}")
-            # Aggregate successful results
-            for result in results:
-                if result is not None:
-                    all_issues.extend(result)
-            logger.info(f"Found {len(all_issues)} issues before deduplication")
+        logger.info(f"Running review modules concurrently: {', '.join(module_names)}...")
+        all_issues = asyncio.run(
+            self._run_review_modules(scopes, repo_path, module_names)
+        )
+        logger.info(f"Found {len(all_issues)} issues before deduplication")
 
         # Deduplicate issues across reviewers (deduplicator checks if enabled internally)
         if len(all_issues) > 1:
@@ -178,7 +198,7 @@ class ReviewPipeline(dspy.Module):
             recommendation = "NEEDS_DISCUSSION" if all_issues else "APPROVE"
         # Collect per-signature statistics
         signature_stats_list = self._collect_signature_stats()
-        
+
         return ReviewResult(
             mr_number=mr.number,
             mr_title=mr.title,
@@ -197,13 +217,13 @@ class ReviewPipeline(dspy.Module):
 
     def _collect_signature_stats(self) -> list[SignatureStatsResult]:
         """Collect statistics from all signatures that executed.
-        
+
         Returns:
             List of SignatureStatsResult for each signature that ran
         """
         stats_list: list[SignatureStatsResult] = []
         all_signature_stats = self.cost_tracker.get_all_signature_stats()
-        
+
         for signature_name, stats in all_signature_stats.items():
             stats_list.append(SignatureStatsResult(
                 name=signature_name,
@@ -212,5 +232,5 @@ class ReviewPipeline(dspy.Module):
                 call_count=stats.call_count,
                 duration_seconds=stats.duration_seconds,
             ))
-        
-        return stats_list
+
+        return stats_list

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/src/codespy/config.py

@@ -330,32 +330,23 @@ class Settings(BaseSettings):
 
     @model_validator(mode="after")
     def sync_llm_settings(self) -> "Settings":
-        """Sync LLM settings between nested and flat fields."""
-        # Sync from nested to flat
-        if self.llm.openai_api_key:
-            self.openai_api_key = self.llm.openai_api_key
-        if self.llm.anthropic_api_key:
-            self.anthropic_api_key = self.llm.anthropic_api_key
-        if self.llm.gemini_api_key:
-            self.gemini_api_key = self.llm.gemini_api_key
-        if self.llm.aws_region:
-            self.aws_region = self.llm.aws_region
-        if self.llm.aws_access_key_id:
-            self.aws_access_key_id = self.llm.aws_access_key_id
-        if self.llm.aws_secret_access_key:
-            self.aws_secret_access_key = self.llm.aws_secret_access_key
-
-        # Sync from flat to nested
-        if self.openai_api_key and not self.llm.openai_api_key:
-            self.llm.openai_api_key = self.openai_api_key
-        if self.anthropic_api_key and not self.llm.anthropic_api_key:
-            self.llm.anthropic_api_key = self.anthropic_api_key
-        if self.gemini_api_key and not self.llm.gemini_api_key:
-            self.llm.gemini_api_key = self.gemini_api_key
-
-        # Sync prompt caching setting
-        self.enable_prompt_caching = self.llm.enable_prompt_caching
+        """Sync LLM settings between nested and flat fields.
 
+        Delegates to LLMConfig.sync_from_flat which enforces:
+        env vars (flat fields from pydantic-settings) > YAML/nested > defaults.
+        """
+        merged = self.llm.sync_from_flat(
+            openai_api_key=self.openai_api_key,
+            anthropic_api_key=self.anthropic_api_key,
+            gemini_api_key=self.gemini_api_key,
+            aws_region=self.aws_region,
+            aws_access_key_id=self.aws_access_key_id,
+            aws_secret_access_key=self.aws_secret_access_key,
+        )
+        # Propagate merged values back to flat fields
+        for field, value in merged.items():
+            if value is not None:
+                setattr(self, field, value)
         return self
 
     @model_validator(mode="after")

{codespy_ai-0.2.0 → codespy_ai-0.2.2}/src/codespy/config_llm.py

@@ -268,4 +268,43 @@ class LLMConfig(BaseModel):
     azure_api_version: str | None = None
 
     # Enable provider-side prompt caching (Anthropic, OpenAI, Bedrock, etc.)
-    enable_prompt_caching: bool = True
+    enable_prompt_caching: bool = True
+
+    def sync_from_flat(
+        self,
+        *,
+        openai_api_key: str | None = None,
+        anthropic_api_key: str | None = None,
+        gemini_api_key: str | None = None,
+        aws_region: str | None = None,
+        aws_access_key_id: str | None = None,
+        aws_secret_access_key: str | None = None,
+    ) -> dict[str, object]:
+        """Sync flat settings into this LLMConfig and return values to propagate back.
+
+        Priority: flat fields (from env vars via pydantic-settings) > nested defaults.
+        Returns dict of field values that should be set on the parent Settings.
+        """
+        # Step 1: Flat → nested (env vars win over nested defaults)
+        if openai_api_key:
+            self.openai_api_key = openai_api_key
+        if anthropic_api_key:
+            self.anthropic_api_key = anthropic_api_key
+        if gemini_api_key:
+            self.gemini_api_key = gemini_api_key
+        if aws_region:
+            self.aws_region = aws_region
+        if aws_access_key_id:
+            self.aws_access_key_id = aws_access_key_id
+        if aws_secret_access_key:
+            self.aws_secret_access_key = aws_secret_access_key
+
+        # Step 2: Return merged values (nested fills gaps where flat is not set)
+        return {
+            "openai_api_key": openai_api_key or self.openai_api_key,
+            "anthropic_api_key": anthropic_api_key or self.anthropic_api_key,
+            "gemini_api_key": gemini_api_key or self.gemini_api_key,
+            "aws_access_key_id": aws_access_key_id or self.aws_access_key_id,
+            "aws_secret_access_key": aws_secret_access_key or self.aws_secret_access_key,
+            "enable_prompt_caching": self.enable_prompt_caching,
+        }