Axom MCP Server

Instructions

Implementation Reference

• src/axom_mcp/handlers/transform.py:25-68 (handler)

  Main handler function 'handle_transform' that executes the axom_mcp_transform tool. It validates input using TransformInput schema, auto-detects input format, parses the input, applies transformation rules, and formats the output.
  async def handle_transform(arguments: Dict[str, Any]) -> str: """Handle axom_mcp_transform tool calls. Args: arguments: Tool arguments containing input data and format parameters Returns: JSON string with transformation result """ # Validate input input_data = TransformInput(**arguments) input_str = input_data.input input_format = input_data.input_format output_format = input_data.output_format rules = input_data.rules or [] template = input_data.template try: # Auto-detect input format if not specified if input_format is None: input_format = _detect_format(input_str) # Parse input to Python object parsed = _parse_input(input_str, input_format) # Apply transformation rules for rule in rules: parsed = _apply_rule(parsed, rule) # Convert to output format result = _format_output(parsed, output_format, template) return json.dumps( { "success": True, "input_format": input_format, "output_format": output_format, "result": result, } ) except Exception as e: logger.error(f"Transform failed: {e}") return json.dumps({"error": str(e)})
• src/axom_mcp/schemas.py:288-312 (schema)

  Input schema definition 'TransformInput' for the axom_mcp_transform tool. Defines required fields (input, output_format) and optional fields (input_format, rules, template, chain) with validation constraints.
  class TransformInput(BaseModel): """Input schema for axom_mcp_transform tool.""" model_config = {"extra": "forbid"} input: str = Field( ..., min_length=1, max_length=10_000_000, description="Input data to transform" ) input_format: Optional[str] = Field( default=None, pattern="^(json|yaml|csv|markdown|code)$", description="Input format (auto-detected if not specified)", ) output_format: str = Field( ..., pattern="^(json|yaml|csv|markdown|code)$", description="Output format" ) rules: Optional[List[Dict[str, Any]]] = Field( default=None, description="Transformation rules" ) template: Optional[str] = Field( default=None, max_length=10000, description="Template for transformation" ) chain: Optional[List[Dict[str, Any]]] = Field( default=None, max_length=10, description="Chain operations after transformation" )
• src/axom_mcp/server.py:329-383 (registration)

  Tool registration for axom_mcp_transform in the TOOLS list. Defines the tool's name, description, input schema, and annotations. The schema specifies all supported formats (json, yaml, csv, markdown, code) and transformation options.
  Tool( name="axom_mcp_transform", description="""Transform data between formats and structures. Supported Formats: - json: JSON objects and arrays - yaml: YAML documents - csv: Comma-separated values - markdown: Markdown documents - code: Source code (with language detection) Transformation Rules: - field_mapping: Rename or restructure fields - filter: Include/exclude specific fields - sort: Sort arrays by field - aggregate: Group and aggregate data Template Support: Use Jinja2 templates for custom output formatting. Chain Support: Use chain parameter to continue processing transformed data.""", inputSchema={ "type": "object", "properties": { "input": {"type": "string", "description": "Input data to transform"}, "input_format": { "type": "string", "enum": ["json", "yaml", "csv", "markdown", "code"], "description": "Input format (auto-detected if not specified)", }, "output_format": { "type": "string", "enum": ["json", "yaml", "csv", "markdown", "code"], "description": "Output format", }, "rules": { "type": "array", "items": {"type": "object"}, "description": "Transformation rules", }, "template": { "type": "string", "description": "Template for transformation", }, "chain": { "type": "array", "items": {"type": "object"}, "description": "Chain operations after transformation", }, }, "required": ["input", "output_format"], }, annotations=TOOL_ANNOTATIONS["transform"], ),
• src/axom_mcp/server.py:498-499 (registration)

  Routing logic that calls handle_transform when the axom_mcp_transform tool is invoked. Maps the tool name to its handler function.
  elif name == "axom_mcp_transform": result = await handle_transform(arguments)
• src/axom_mcp/handlers/transform.py:71-115 (helper)

  Helper function _detect_format that auto-detects input format based on content patterns. Checks for JSON, YAML, CSV, markdown, and defaults to code format using regex heuristics.
  def _detect_format(input_str: str) -> str: """Auto-detect input format based on content.""" stripped = input_str.strip() # Check for JSON if stripped.startswith("{") or stripped.startswith("["): try: json.loads(stripped) return "json" except json.JSONDecodeError: pass # Check for YAML (basic heuristics) yaml_indicators = [ r"^\w+:\s*", # key: value r"^-\s+", # - list item (can be at start of line) r"^\s+-\s+", # - list item (with indentation) r"^---\s*$", # YAML document separator ] for pattern in yaml_indicators: if re.search(pattern, stripped, re.MULTILINE): return "yaml" # Check for CSV lines = stripped.split("\n") if len(lines) > 1: # Check if all lines have similar number of commas comma_counts = [line.count(",") for line in lines[:5]] if len(set(comma_counts)) == 1 and comma_counts[0] > 0: return "csv" # Check for Markdown md_indicators = [ r"^#+\s+", # Headers r"^\*\s+", # Unordered lists r"^\d+\.\s+", # Ordered lists r"\[.*\]\(.*\)", # Links r"```", # Code blocks ] for pattern in md_indicators: if re.search(pattern, stripped, re.MULTILINE): return "markdown" # Default to code return "code"