Feature: override default prompt with custom supporting placeholders

CLAUDE.md

@@ -11,6 +11,9 @@ LLM-powered code review system with three integration modes:
 
 Uses OpenAI-compatible LLM API (configurable model and endpoint).
 
+### Supported Languages
+Python, JavaScript, TypeScript, JSX, TSX, Java, C, C++, Go
+
 ## Common Commands
 
 ```bash
@@ -36,6 +39,11 @@ python review.py --test-connection          # Test LLM API connectivity
 # Install git hooks
 python install_hooks.py
 
+# Code formatting and linting
+ruff format .                               # Format all Python files
+ruff check .                                # Lint all Python files
+ruff check --fix .                          # Auto-fix linting issues
+
 # Monitoring
 python monitor.py health                    # System health check
 python monitor.py report --days 7           # Usage report
@@ -88,3 +96,4 @@ static_analyzer.py  Fallback security analysis when LLM unavailable
 - **All code comments must be in English**
 - Communication with users can be in Ukrainian
 - Uses fish shell for environment configuration
+- Code formatted with `ruff format` and linted with `ruff check`

config.py

@@ -21,6 +21,9 @@ class ReviewConfig:
             "api_key_env": "LLM_API_KEY",
             "timeout": 30,
             "max_retries": 3,
+            "max_tokens_per_request": 4096,
+            "token_limit_strategy": "chunk",  # "truncate", "chunk", or "skip"
+            "chars_per_token": 4,  # Character-to-token ratio for estimation
         },
         "prompt": {
             "custom_prompt": None,
@@ -161,6 +164,65 @@ def get_model(self) -> str:
         logger.debug("Using model from config: %s", config_model)
         return config_model
 
+    def get_max_tokens(self) -> int:
+        """Get max tokens per request. Environment variable takes precedence."""
+        env_value = os.getenv("LLM_MAX_TOKENS_PER_REQUEST")
+        if env_value:
+            try:
+                value = int(env_value)
+                logger.debug(
+                    "Using max tokens from LLM_MAX_TOKENS_PER_REQUEST: %d", value
+                )
+                return value
+            except ValueError:
+                logger.warning(
+                    "Invalid LLM_MAX_TOKENS_PER_REQUEST value: %s. Using config.",
+                    env_value,
+                )
+
+        config_value = self.get("llm.max_tokens_per_request", 4096)
+        logger.debug("Using max tokens from config: %d", config_value)
+        return config_value
+
+    def get_token_limit_strategy(self) -> str:
+        """Get token limit strategy. Environment variable takes precedence."""
+        env_value = os.getenv("LLM_TOKEN_LIMIT_STRATEGY")
+        valid_strategies = ("truncate", "chunk", "skip")
+
+        if env_value:
+            if env_value in valid_strategies:
+                logger.debug(
+                    "Using token limit strategy from LLM_TOKEN_LIMIT_STRATEGY: %s",
+                    env_value,
+                )
+                return env_value
+            else:
+                logger.warning(
+                    "Invalid LLM_TOKEN_LIMIT_STRATEGY value: %s. Must be one of %s. Using config.",
+                    env_value,
+                    valid_strategies,
+                )
+
+        config_value = self.get("llm.token_limit_strategy", "chunk")
+        if config_value not in valid_strategies:
+            logger.warning(
+                "Invalid token_limit_strategy in config: %s. Using 'chunk'.",
+                config_value,
+            )
+            return "chunk"
+
+        logger.debug("Using token limit strategy from config: %s", config_value)
+        return config_value
+
+    def get_chars_per_token(self) -> int:
+        """Get character-to-token ratio for estimation."""
+        config_value = self.get("llm.chars_per_token", 4)
+        if config_value < 1:
+            logger.warning("chars_per_token must be >= 1. Using default 4.")
+            return 4
+        logger.debug("Using chars_per_token: %d", config_value)
+        return config_value
+
     def is_file_supported(self, file_path: str) -> bool:
         """Check if file extension is supported for review."""
         path = Path(file_path)

custom_prompt_example.txt

@@ -0,0 +1,58 @@
+Custom Prompt Template Example
+==============================
+
+When using "custom_prompt" in your review_config.json, you can use
+the following placeholders:
+
+  {diff_content}            - The git diff to review
+  {custom_critical_rules}   - Formatted list of custom critical rules
+  {custom_warnings}         - Formatted list of custom warnings
+  {custom_suggestions}      - Formatted list of custom suggestions
+  {additional_instructions} - Additional instructions text
+
+All placeholders are optional - use only the ones you need.
+
+Example Configuration
+---------------------
+{
+  "prompt": {
+    "custom_prompt": "You are a code reviewer.\n\nCRITICAL:\n{custom_critical_rules}\n\nWARNINGS:\n{custom_warnings}\n\nSUGGESTIONS:\n{custom_suggestions}\n\n{additional_instructions}\n\nReview:\n{diff_content}\n\nRespond with CRITICAL:, WARNING:, or SUGGESTION: prefixes.",
+    "custom_critical_rules": ["memory leaks", "race conditions"],
+    "custom_warnings": ["deprecated API usage"],
+    "custom_suggestions": ["consider async patterns"],
+    "additional_instructions": "Focus on thread safety"
+  }
+}
+
+Minimal Example (diff only)
+---------------------------
+{
+  "prompt": {
+    "custom_prompt": "Review this code for security issues:\n\n{diff_content}\n\nRespond with CRITICAL:, WARNING:, or SUGGESTION: prefixes."
+  }
+}
+
+Full Custom Prompt Template
+---------------------------
+You are a security-focused code reviewer. Analyze the following changes.
+
+CRITICAL ISSUES (block commit):
+{custom_critical_rules}
+
+WARNINGS (allow but flag):
+{custom_warnings}
+
+SUGGESTIONS (improvements):
+{custom_suggestions}
+
+{additional_instructions}
+
+Format your response as:
+CRITICAL: [issue description]
+WARNING: [issue description]
+SUGGESTION: [suggestion]
+
+If no issues found for a category, respond "NONE".
+
+Changes to review:
+{diff_content}

review.py

@@ -214,7 +214,9 @@ def _format_text_output(self, result: ReviewResult, verbose: bool = False) -> st
         lines = []
 
         # Status indicator
-        if result.status == "model_unavailable":
+        if result.status == "skipped":
+            lines.append("⏭️ Review Skipped (token limit exceeded)")
+        elif result.status == "model_unavailable":
             lines.append("🔴 LLM Model Unavailable")
         elif result.critical_issues:
             lines.append("❌ Critical Issues Found")
@@ -252,13 +254,19 @@ def _format_text_output(self, result: ReviewResult, verbose: bool = False) -> st
             lines.append(f"   • Review Status: {result.status}")
             if result.fallback_used:
                 lines.append("   • Fallback Analysis: Yes")
+            if result.total_chunks > 0:
+                lines.append(
+                    f"   • Chunks Reviewed: {result.chunks_reviewed}/{result.total_chunks}"
+                )
             lines.append("")
 
         return "\n".join(lines)
 
     def _get_exit_code(self, result: ReviewResult, args=None) -> int:
         """Get appropriate exit code based on results."""
-        if result.status == "model_unavailable":
+        if result.status == "skipped":
+            return 5  # Review skipped due to token limit
+        elif result.status == "model_unavailable":
             return 3  # Model unavailable, allow commit with warning
         elif result.critical_issues:
             return 1  # Critical issues, block commit

review_config.json

@@ -5,7 +5,10 @@
     "api_key_env": "LLM_API_KEY",
     "timeout": 30,
     "max_retries": 3,
-    "fallback_model": "anthropic/claude-haiku"
+    "fallback_model": "anthropic/claude-haiku",
+    "max_tokens_per_request": 4096,
+    "token_limit_strategy": "chunk",
+    "chars_per_token": 4
   },
   "prompt": {
     "custom_prompt": null,

review_config_example.json

@@ -32,6 +32,13 @@
       "*.spec.js"
     ]
   },
+  "prompt": {
+    "custom_prompt": null,
+    "custom_critical_rules": ["memory leaks", "race conditions"],
+    "custom_warnings": ["deprecated API usage"],
+    "custom_suggestions": ["consider async patterns"],
+    "additional_instructions": "Focus on thread safety"
+  },
   "output": {
     "format": "text",
     "show_context": true,