Code AnalyzerBLOCK
A Model Context Protocol (MCP) server implementation for comprehensive code analysis. This tool integrates with Claude Desktop to provide code analysis capabilities through natural language interactions.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The main purpose of the MCP tool is to adapt changes to the project intelligently. For instance, when a code modification or structural adjustment is needed, it aims to ensure that other related usages are also updated accordingly.
Currently, the project has a lot of bugs, and the tools are not functioning as expected sometimes. Due to work commitments, I can’t dedicate much time to it, but I’ll try to fix the issues and bring it to a working state whenever I find the time.
⚠️ WARNING: BACKUP RECOMMENDED This tool performs file operations that could potentially modify or delete files. Always backup your codebase before using any modification features.
Overview
The Model Context Protocol (MCP) Code Analyzer is a comprehensive code analysis and management tool that integrates with Claude Desktop. It analyzes code at both project and file levels, providing insights for needed changes and project understanding.
Prerequisites
- Python 3.10 or later
- Windows (Linux/MacOS support not yet tested)
Installation
# Clone the repository git clone https://github.com/[your-username]/mcp-code-analyzer.git cd mcp-code-analyzer # Install dependencies and package pip install -e .
Claude Desktop Integration
Configuration
- Access Claude Desktop config:
Win + R → %AppData%\Claude\
- Create/edit
claude_desktop_config.json:
{
"globalShortcut": "Ctrl+Space",
"mcpServers": {
"code-analyzer": {
"command": "python",
"args": [
"-m",
"mcp_code_analyzer",
"--analyze-paths",
"C:\\Projects\\path1"
],
"type": "module"
}
}
}Path Configuration
- Multiple project paths can be specified in configuration
- Additional paths or files can be analyzed via chat messages
- No limit on number of analyzable paths/files
🛠️ Tools and Status
Working Tools
- ✅ Project Analysis
57336eda3f59OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add mcp-code-analyzer -- uvx mcp-code-analyzer
{
"mcpServers": {
"mcp-code-analyzer": {
"command": "uvx",
"args": [
"mcp-code-analyzer"
]
}
}
}Exposed tools (18)
17 read · 1 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
analyze_code_structure | read | Analyze code structure and architecture |
analyze_imports | write | Analyze import statements and dependencies |
analyze_project_statistics | read | Project-wide statistics and metrics |
analyze_project_structure | read | Directory structure and organization analysis with tree view |
analyze_project_technology | read | Detect and analyze used technologies and frameworks |
check_syntax | read | Advanced syntax checking and analysis |
code_modifier | read | Safe code modification with impact analysis |
file_operations | read | File operations with MCP support |
find_code_patterns | read | Detect code patterns and anti-patterns |
find_pattern_usages | read | Find pattern occurrences and analyze usage |
find_references | read | Find code references |
manage_changes | read | Manage code changes and their application |
preview_changes | read | Preview code changes |
search_code | read | Search code with pattern matching |
search_content | read | Search within file contents |
search_files | read | Advanced file search capabilities |
validate_code | read | Validate code quality and standards |
version_control | read | Advanced version control and history management |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- declared (3 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (5)
r"eval\(": "Use of eval() is dangerous",r"exec\(": "Use of exec() is dangerous",return hashlib.md5(content.encode()).hexdigest()
return hashlib.md5(unique_data.encode()).hexdigest()[:12]
chardet, astroid, pydantic, radon, networkx, mcp
Gates applied: no_behavioural_pass.
57336eda3f59full audit observations/trust-audit/mcp-server/emiryasar__code-analyzer-1.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 57336eda3f59 | BLOCK | D | 69 | first audit |
Questions
What is the Code Analyzer MCP server?
A Model Context Protocol (MCP) server implementation for comprehensive code analysis. This tool integrates with Claude Desktop to provide code analysis capabilities through natural language interactions.
What tools does Code Analyzer expose?
18 in total: 17 read-only, 1 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Code Analyzer safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does Code Analyzer need?
No credential environment variables were found in its source, so it appears to need none.
How does Code Analyzer run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as mcp-code-analyzer.
How current is this page?
The grade is for one exact copy of the source (57336eda3f59), read on 2026-10-08. The repository is watched and re-audited when it changes.