Multi-protocol MCP server with support for Aider, Code2LLM, and various standard MCP tools (Audio, Brave Search, Browser, Docker-AI, DuckDuckGo, Fetch, Filesystem, Git, GitHub, Image, Jira, Kubernetes, Mail, MySQL, PDF, Playwright)
Critical failures across all dimensions. 38 tools are defined only in Dockerfiles and wrapper containers with NO visible source code showing actual tool registration, schemas, or parameter definitions. Tool definitions are inferred from brief descriptions in the provided metadata, not from actual MCP server implementations. No input schemas are visible in the source. Descriptions are minimal (10-50 chars) and lack actionable context. No error handling, no output schema documentation, no parameter validation guidance. This is a Docker orchestration layer, not a functional MCP server implementation.
Edit file using Aider AI assistant
Synthesize text to speech
Transcribe audio files to text using speech-to-text
Search images using Brave Search API
Search news using Brave Search API
Search the web using Brave Search API
Click an element in the browser
Navigate to a URL in the browser
NO INPUT SCHEMAS VISIBLE. All 38 tools lack visible JSON Schema definitions in source code. Tool definitions are inferred from Dockerfile labels and metadata only, not from actual MCP server code.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 33 | 2026-07-28+ | v2 |
Analyze code structure and complexity using code2llm
Execute command in Docker container with AI assistance
List Docker containers with AI analysis
Search the web using DuckDuckGo (no API key required)
Fetch content from a URL
Delete file from filesystem
Read file from filesystem
Write file to filesystem
Clone a git repository
Get git diff between commits
Get git commit log
Create a GitHub issue
Create a GitHub pull request
List GitHub repositories
Analyze image content
Crop an image
Resize an image
Create a Jira issue
List Jira issues
Execute command in Kubernetes pod
Get logs from Kubernetes pod
List Kubernetes pods
Send email via SMTP
List MySQL database tables
Execute MySQL query
Extract text from PDF file
Parse PDF structure and metadata
Fill form field using Playwright
Navigate to URL using Playwright
Take screenshot with Playwright
TOOL DEFINITIONS INFERRED, NOT VISIBLE. Source code shows only Dockerfiles wrapping external MCP servers (mcp/audio, mcp/brave-search, etc.) and two custom servers (aider_mcp_server.py, code2llm_mcp_server.py). The actual tool registration code is not provided.
MINIMAL DESCRIPTIONS (10-50 chars). All tool descriptions are under 100 characters and lack actionable context. Examples: 'Transcribe audio files to text using speech-to-text' (50 chars) does not explain WHEN to use it, what formats are supported, or what the output contains.
NO PARAMETER DESCRIPTIONS. Input parameters (file_path, instruction, language, voice, query, etc.) lack descriptions explaining what they control, expected format, or constraints.
NO OUTPUT SCHEMA DOCUMENTATION. No tool documents what fields it returns, data types, or structure. LLMs cannot plan downstream tool calls or extract required data without knowing output shape.