jarvis-ai-assistant 0.1.103__tar.gz → 0.1.104__tar.gz

{jarvis_ai_assistant-0.1.103/src/jarvis_ai_assistant.egg-info → jarvis_ai_assistant-0.1.104}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: jarvis-ai-assistant
-Version: 0.1.103
+Version: 0.1.104
 Summary: Jarvis: An AI assistant that uses tools to interact with the system
 Home-page: https://github.com/skyfireitdiy/Jarvis
 Author: skyfire
@@ -153,20 +153,18 @@ Jarvis supports configuration through environment variables that can be set in t
 
 ## 🎯 Usage
 
-### Main Assistant
+### Code Modification
 ```bash
+# Using main agent
 jarvis
+
+# Using code agent directly
+jarvis-code-agent
 ```
 
-### Codebase Search
+### Codebase Query
 ```bash
-# Generate codebase index
-jarvis-codebase generate
-
-# Search similar code
-jarvis-codebase search "your search query"
-
-# Ask questions about codebase
+# Ask questions about the codebase
 jarvis-codebase ask "your question"
 ```
 
@@ -175,71 +173,98 @@ jarvis-codebase ask "your question"
 # Build document index
 jarvis-rag --dir /path/to/documents --build
 
-# Search documents
-jarvis-rag --query "your search query"
+# Ask questions about documents
+jarvis-rag --query "your question"
+```
+
+### Smart Shell
+```bash
+# Using full name
+jarvis-smart-shell "describe what you want to do"
+
+# Using shorthand
+jss "describe what you want to do"
 ```
 
-### Search Tool
+### Development Tools
 ```bash
-# Basic search
-jarvis-search "your query"
+# Manage git commits
+jarvis-git-commit
 
-# Show only URLs
-jarvis-search "your query" --url-only
+# Generate and manage ctags
+jarvis-ctags
 
-# Limit results
-jarvis-search "your query" --max 3
+# Manage AI platforms
+jarvis-platform
 ```
 
-## 🛠️ Tools
+Each command supports `--help` flag for detailed usage information:
+```bash
+jarvis --help
+jarvis-code-agent --help
+jarvis-codebase --help
+jarvis-rag --help
+jarvis-smart-shell --help
+jarvis-platform --help
+jarvis-git-commit --help
+jarvis-ctags --help
+```
 
+## 🛠️ Tools
 
 ### Built-in Tools
 
 | Tool | Description |
 |------|-------------|
+| read_code | Read code files with line numbers and range support |
 | execute_shell | Execute system commands and capture output |
-| file_operation | File operations (read/write/append/delete) |
-| generate_tool | AI-powered tool generation and integration |
-| methodology | Experience accumulation and methodology management |
-| create_sub_agent | Create specialized sub-agents for specific tasks |
-| coder | Automatic code modification and generation tool |
-| codebase | Codebase management and search tool |
+| search | Web search for development related queries |
+| ask_user | Interactive user input collection |
+| ask_codebase | Intelligent codebase querying and analysis |
+| code_review | Automated code review with multi-dimensional analysis |
+| file_operation | Basic file operations (read/exists) |
+| git_commiter | Automated git commit handling |
 
 ### Tool Locations
 - Built-in tools: `src/jarvis/tools/`
 - User tools: `~/.jarvis/tools/`
 
-
 ### Key Features
 
-#### 1. Self-Extending Capabilities
-- Tool generation through natural language description
-- Automatic code generation and integration
-- Dynamic capability expansion through sub-agents
-- Automatic code modification with version control
-- Codebase indexing and semantic search
-
-#### 2. Methodology Learning
-- Automatic experience accumulation from interactions
-- Pattern recognition and methodology extraction
-- Continuous refinement through usage
-- Code modification history tracking
-- Codebase analysis and documentation generation
-
-#### 3. Adaptive Problem Solving
-- Context-aware sub-agent creation
-- Dynamic tool composition
-- Learning from execution feedback
+#### 1. Code Intelligence
+- Smart file selection and analysis based on requirements
+- Semantic codebase search and querying
+- Efficient handling of large files with context-aware reading
+- Precise patch-based code modifications
+- Automated git commit management
+
+#### 2. Multi-Model Architecture
+- Support for multiple AI platforms (Kimi/OpenAI/AI8/OYI/Ollama)
+- Platform-specific optimizations for different tasks
+- Specialized models for code generation, thinking, and general tasks
+- Streaming response support for better interaction
+- Automatic model fallback and retry mechanisms
+
+#### 3. RAG Capabilities
+- Document indexing and semantic search
+- Smart context management for large documents
+- Automatic file change detection
+- Efficient caching mechanisms
+- Multi-format document support
+
+#### 4. Development Tools
+- Interactive shell command generation
+- Code review with multi-dimensional analysis
 - Codebase-aware problem solving
-- Multi-model collaboration for complex tasks
-
-#### 4. Code Intelligence
-- Automatic codebase indexing
-- Semantic code search
-- Code modification with git integration
-- Code analysis and documentation
-- Multi-model code generation
+- File operations with safety checks
+- Progress tracking and error handling
+
+#### 5. User Experience
+- Beautiful console output with color support
+- Interactive multi-line input
+- Progress indicators for long operations
+- Clear error messages and handling
+- Context-aware response formatting
 
 ## 🎯 Extending Jarvis
 

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/README.md

@@ -83,20 +83,18 @@ Jarvis supports configuration through environment variables that can be set in t
 
 ## 🎯 Usage
 
-### Main Assistant
+### Code Modification
 ```bash
+# Using main agent
 jarvis
+
+# Using code agent directly
+jarvis-code-agent
 ```
 
-### Codebase Search
+### Codebase Query
 ```bash
-# Generate codebase index
-jarvis-codebase generate
-
-# Search similar code
-jarvis-codebase search "your search query"
-
-# Ask questions about codebase
+# Ask questions about the codebase
 jarvis-codebase ask "your question"
 ```
 
@@ -105,71 +103,98 @@ jarvis-codebase ask "your question"
 # Build document index
 jarvis-rag --dir /path/to/documents --build
 
-# Search documents
-jarvis-rag --query "your search query"
+# Ask questions about documents
+jarvis-rag --query "your question"
+```
+
+### Smart Shell
+```bash
+# Using full name
+jarvis-smart-shell "describe what you want to do"
+
+# Using shorthand
+jss "describe what you want to do"
 ```
 
-### Search Tool
+### Development Tools
 ```bash
-# Basic search
-jarvis-search "your query"
+# Manage git commits
+jarvis-git-commit
 
-# Show only URLs
-jarvis-search "your query" --url-only
+# Generate and manage ctags
+jarvis-ctags
 
-# Limit results
-jarvis-search "your query" --max 3
+# Manage AI platforms
+jarvis-platform
 ```
 
-## 🛠️ Tools
+Each command supports `--help` flag for detailed usage information:
+```bash
+jarvis --help
+jarvis-code-agent --help
+jarvis-codebase --help
+jarvis-rag --help
+jarvis-smart-shell --help
+jarvis-platform --help
+jarvis-git-commit --help
+jarvis-ctags --help
+```
 
+## 🛠️ Tools
 
 ### Built-in Tools
 
 | Tool | Description |
 |------|-------------|
+| read_code | Read code files with line numbers and range support |
 | execute_shell | Execute system commands and capture output |
-| file_operation | File operations (read/write/append/delete) |
-| generate_tool | AI-powered tool generation and integration |
-| methodology | Experience accumulation and methodology management |
-| create_sub_agent | Create specialized sub-agents for specific tasks |
-| coder | Automatic code modification and generation tool |
-| codebase | Codebase management and search tool |
+| search | Web search for development related queries |
+| ask_user | Interactive user input collection |
+| ask_codebase | Intelligent codebase querying and analysis |
+| code_review | Automated code review with multi-dimensional analysis |
+| file_operation | Basic file operations (read/exists) |
+| git_commiter | Automated git commit handling |
 
 ### Tool Locations
 - Built-in tools: `src/jarvis/tools/`
 - User tools: `~/.jarvis/tools/`
 
-
 ### Key Features
 
-#### 1. Self-Extending Capabilities
-- Tool generation through natural language description
-- Automatic code generation and integration
-- Dynamic capability expansion through sub-agents
-- Automatic code modification with version control
-- Codebase indexing and semantic search
-
-#### 2. Methodology Learning
-- Automatic experience accumulation from interactions
-- Pattern recognition and methodology extraction
-- Continuous refinement through usage
-- Code modification history tracking
-- Codebase analysis and documentation generation
-
-#### 3. Adaptive Problem Solving
-- Context-aware sub-agent creation
-- Dynamic tool composition
-- Learning from execution feedback
+#### 1. Code Intelligence
+- Smart file selection and analysis based on requirements
+- Semantic codebase search and querying
+- Efficient handling of large files with context-aware reading
+- Precise patch-based code modifications
+- Automated git commit management
+
+#### 2. Multi-Model Architecture
+- Support for multiple AI platforms (Kimi/OpenAI/AI8/OYI/Ollama)
+- Platform-specific optimizations for different tasks
+- Specialized models for code generation, thinking, and general tasks
+- Streaming response support for better interaction
+- Automatic model fallback and retry mechanisms
+
+#### 3. RAG Capabilities
+- Document indexing and semantic search
+- Smart context management for large documents
+- Automatic file change detection
+- Efficient caching mechanisms
+- Multi-format document support
+
+#### 4. Development Tools
+- Interactive shell command generation
+- Code review with multi-dimensional analysis
 - Codebase-aware problem solving
-- Multi-model collaboration for complex tasks
-
-#### 4. Code Intelligence
-- Automatic codebase indexing
-- Semantic code search
-- Code modification with git integration
-- Code analysis and documentation
-- Multi-model code generation
+- File operations with safety checks
+- Progress tracking and error handling
+
+#### 5. User Experience
+- Beautiful console output with color support
+- Interactive multi-line input
+- Progress indicators for long operations
+- Clear error messages and handling
+- Context-aware response formatting
 
 ## 🎯 Extending Jarvis
 

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jarvis-ai-assistant"
-version = "0.1.103"
+version = "0.1.104"
 description = "Jarvis: An AI assistant that uses tools to interact with the system"
 readme = "README.md"
 authors = [{ name = "Your Name", email = "your.email@example.com" }]
@@ -58,6 +58,6 @@ jarvis-rag = "jarvis.jarvis_rag.main:main"
 jarvis-smart-shell = "jarvis.jarvis_smart_shell.main:main"
 jss = "jarvis.jarvis_smart_shell.main:main"
 jarvis-platform = "jarvis.jarvis_platform.main:main"
-jarvis-gh = "jarvis.jarvis_github.main:main"
 jarvis-git-commit = "jarvis.tools.git_commiter:main"
 jarvis-ctags = "jarvis.tools.create_ctags_agent:main"
+jarvis-code-review = "jarvis.tools.code_review:main"

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name="jarvis-ai-assistant",
-    version="0.1.103",
+    version="0.1.104",
     author="skyfire",
     author_email="skyfireitdiy@hotmail.com",
     description="An AI assistant that uses various tools to interact with the system",
@@ -43,9 +43,9 @@ setup(
             "jarvis-smart-shell=jarvis.jarvis_smart_shell.main:main",
             "jss=jarvis.jarvis_smart_shell.main:main",
             "jarvis-platform=jarvis.jarvis_platform.main:main",
-            "jarvis-gh=jarvis.jarvis_github.main:main",
             "jarvis-git-commit=jarvis.tools.git_commiter:main",
             "jarvis-ctags=jarvis.tools.create_ctags_agent:main",
+            "jarvis-code-review=jarvis.tools.code_review:main",
         ],
     },
     python_requires=">=3.8",

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/__init__.py

@@ -1,3 +1,3 @@
 """Jarvis AI Assistant"""
 
-__version__ = "0.1.103" 
+__version__ = "0.1.104" 

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_code_agent/file_select.py

@@ -3,7 +3,7 @@ import re
 from typing import Dict, List
 from prompt_toolkit import PromptSession
 from prompt_toolkit.completion import WordCompleter, Completer, Completion
-from jarvis.utils import OutputType, PrettyOutput, get_single_line_input
+from jarvis.utils import OutputType, PrettyOutput, get_single_line_input, user_confirm
 
 
 def _parse_file_selection(input_str: str, max_index: int) -> List[int]:
@@ -135,11 +135,10 @@ def select_files(related_files: List[str], root_dir: str) -> List[str]:
     for i, file in enumerate(related_files, 1):
         output += f"[{i}] {file}\n"
 
-    PrettyOutput.print(output, OutputType.INFO)
+    PrettyOutput.print(output, OutputType.INFO, lang="markdown")
     
     # Ask the user if they need to adjust the file list
-    user_input = get_single_line_input("Do you need to adjust the file list? (y/n) [n]").strip().lower() or 'n'
-    if user_input == 'y':
+    if user_confirm("Do you need to adjust the file list?", False):
         # Let the user select files
         numbers = get_single_line_input("Please enter the file numbers to include (support: 1,3-6 format, press Enter to keep the current selection)").strip()
         if numbers:
@@ -150,8 +149,7 @@ def select_files(related_files: List[str], root_dir: str) -> List[str]:
                 PrettyOutput.print("No valid files selected, keep the current selection", OutputType.WARNING)
     
     # Ask if they need to supplement files
-    user_input = get_single_line_input("Do you need to supplement other files? (y/n) [n]").strip().lower() or 'n'
-    if user_input == 'y':
+    if user_confirm("Do you need to supplement other files?", False):
         # Create file completion session
         session = PromptSession(
             completer=_get_file_completer(root_dir),

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_code_agent/patch.py

@@ -108,6 +108,8 @@ def handle_commit_workflow()->bool:
     Returns:
         tuple[bool, str, str]: (continue_execution, commit_id, commit_message)
     """
+    diff = os.popen("git diff HEAD").read()
+    PrettyOutput.print(diff, OutputType.CODE, lang="diff")
     if not user_confirm("Do you want to commit the code?", default=True):
         os.system("git reset HEAD")
         os.system("git checkout -- .")

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_code_agent/relevant_files.py

@@ -1,6 +1,3 @@
-
-
-
 import os
 import re
 from typing import List
@@ -27,19 +24,53 @@ def find_relevant_files(user_input: str, root_dir: str) -> List[str]:
         find_file_tool_registry.use_tools(["read_code", "execute_shell"])
         find_file_agent = Agent(
             system_prompt="""You are a file agent, you are responsible for finding files related to the user's requirement.
-        You can use `read_code` tool to read the code and analyze the code, and `execute_shell` tool to execute shell command(such as `grep/find/ls/git/ctags`) to find files.
 
-        IMPORTANT:
-        - Only provide the file path, do not provide any other information.
-        - If you can't find the file, please provide empty list.
-        - Don't modify the code, just find related files.
+        SEARCH STRATEGY:
+        1. First Pass - Quick Search:
+           - Use `execute_shell` with git grep/find to locate potential files
+           - Search for key terms, function names, and relevant patterns
+           - Example: execute_shell("git grep -l 'search_term'")
+
+        2. Content Analysis:
+           - For each potential file, analyze its content
+           - Follow the file reading guidelines for large files
+           - Look for:
+             * Direct matches to requirement terms
+             * Related functionality
+             * Imported/referenced files
+             * Test files for modified code
+
+        FILE READING GUIDELINES:
+        1. For Large Files (>200 lines):
+           - Do NOT read the entire file at once
+           - First use grep/ctags to locate relevant sections
+           - Then read specific sections with context
+           - Example:
+             * execute_shell("grep -n 'function_name' path/to/file")
+             * read_code("path/to/file", start_line=found_line-10, end_line=found_line+20)
+
+        2. For Small Files:
+           - Can read entire file directly
+
+        IMPORTANT RULES:
+        - Only return files that are DIRECTLY related to the requirement
+        - Exclude false positives and loosely related files
+        - If a file only contains imports/references, don't include it
+        - Include both implementation and test files when relevant
+        - If unsure about a file, use grep/read_code to verify relevance
+        - Return empty list if no truly relevant files are found
+        - Do NOT modify any code, only find files
+
+        OUTPUT FORMAT:
+        - Only provide file paths in the specified YAML format
+        - No additional explanations or comments
         """, 
             name="FindFileAgent", 
             is_sub_agent=True,
             tool_registry=find_file_tool_registry, 
             platform=PlatformRegistry().get_normal_platform(),
             auto_complete=True,
-            summary_prompt="""Please provide the file path as this format(yaml list), if you can't find the file, please provide empty list:
+            summary_prompt="""Please provide the file paths as YAML list:
             <FILE_PATH>
             - file_path1
             - file_path2

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_codebase/main.py

@@ -9,7 +9,7 @@ from jarvis.models.registry import PlatformRegistry
 import concurrent.futures
 from threading import Lock
 from concurrent.futures import ThreadPoolExecutor
-from jarvis.utils import OutputType, PrettyOutput, find_git_root, get_file_md5, get_max_context_length, get_single_line_input, get_thread_count, load_embedding_model, load_rerank_model
+from jarvis.utils import OutputType, PrettyOutput, find_git_root, get_file_md5, get_max_context_length, get_single_line_input, get_thread_count, load_embedding_model, load_rerank_model, user_confirm
 from jarvis.utils import init_env
 import argparse
 import pickle
@@ -443,16 +443,9 @@ Content: {content}
 
                 # If force is True, continue directly
                 if not force:
-                    # Ask the user whether to continue
-                    while True:
-                        response = get_single_line_input("Rebuild the index? [y/N]").lower().strip()
-                        if response in ['y', 'yes']:
-                            break
-                        elif response in ['', 'n', 'no']:
-                            PrettyOutput.print("Cancel rebuilding the index", output_type=OutputType.INFO)
-                            return
-                        else:
-                            PrettyOutput.print("Please input y or n", output_type=OutputType.WARNING)
+                    if not user_confirm("Rebuild the index?", False):
+                        PrettyOutput.print("Cancel rebuilding the index", output_type=OutputType.INFO)
+                        return
                 
                 # Clean deleted files
                 for file_path in files_to_delete:
@@ -718,7 +711,7 @@ Please output 3 expressions directly, separated by two line breaks, without numb
             message = "Found related files:\n"
             for path, score, _ in initial_results:
                 message += f"File: {path} Similarity: {score:.3f}\n"
-            PrettyOutput.print(message.rstrip(), output_type=OutputType.INFO)
+            PrettyOutput.print(message.rstrip(), output_type=OutputType.INFO, lang="markdown")
                 
             # Reorder the preliminary results
             return self.pick_results(query, [path for path, _, _ in initial_results])
@@ -737,7 +730,7 @@ Please output 3 expressions directly, separated by two line breaks, without numb
         message = "Found related files:\n"
         for path in results:
             message += f"File: {path}\n"
-        PrettyOutput.print(message.rstrip(), output_type=OutputType.SUCCESS)
+        PrettyOutput.print(message.rstrip(), output_type=OutputType.SUCCESS, lang="markdown")
         
         prompt = f"""You are a code expert, please answer the user's question based on the following file information:
 """
@@ -862,8 +855,8 @@ def main():
             
         output = "Search Results:\n"
         for path in results:
-            output += f"""{path}\n"""
-        PrettyOutput.print(output, output_type=OutputType.INFO)
+            output += f"""- {path}\n"""
+        PrettyOutput.print(output, output_type=OutputType.INFO, lang="markdown")
 
     elif args.command == 'ask':            
         response = codebase.ask_codebase(args.question, args.top_k)

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_platform/main.py

@@ -29,7 +29,7 @@ def list_platforms():
                         output += f"  • {model_name} - {description}\n"
                     else:
                         output += f"  • {model_name}\n"
-                PrettyOutput.print(output, OutputType.SUCCESS)
+                PrettyOutput.print(output, OutputType.SUCCESS, lang="markdown")
             else:
                 PrettyOutput.print("  • No available model information", OutputType.WARNING)
                 

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/jarvis_rag/main.py

@@ -699,7 +699,7 @@ Content: {doc.content}
                 output = f"""File: {doc.metadata['file_path']}\n"""
                 output += f"""Fragment {doc.metadata['chunk_index'] + 1}/{doc.metadata['total_chunks']}\n"""
                 output += f"""Content:\n{doc.content}\n"""
-                PrettyOutput.print(output, output_type=OutputType.INFO)
+                PrettyOutput.print(output, output_type=OutputType.INFO, lang="markdown")
 
             # Build base prompt
             base_prompt = f"""Please answer the user's question based on the following document fragments. If the document content is not sufficient to answer the question completely, please clearly indicate.
@@ -792,7 +792,7 @@ def main():
                     output = f"""File: {doc.metadata['file_path']}\n"""
                     output += f"""Fragment {doc.metadata['chunk_index'] + 1}/{doc.metadata['total_chunks']}\n"""
                     output += f"""Content:\n{doc.content}\n"""
-                    PrettyOutput.print(output, output_type=OutputType.INFO)
+                    PrettyOutput.print(output, output_type=OutputType.INFO, lang="markdown")
                 return 0
 
             if args.ask:

{jarvis_ai_assistant-0.1.103 → jarvis_ai_assistant-0.1.104}/src/jarvis/tools/ask_codebase.py

@@ -87,13 +87,10 @@ def main():
     })
     
     if result["success"]:
-        print(result["stdout"])
+        PrettyOutput.print(result["stdout"], OutputType.INFO, lang="markdown")
     else:
         PrettyOutput.print(result["stderr"], OutputType.ERROR)
-        return 1
         
-    return 0
-
 
 if __name__ == "__main__":
     main()