ai_edit
Edit images using AI with natural language prompts. Supports single and multi-image editing for visual content modification.
Instructions
Edit images using AI with natural language prompts. Supports single and multi-image editing. Cost: $0.15. Requires Tier 1+.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| media_url | Yes | URL of the primary image to edit | |
| prompt | Yes | Natural language instruction for editing | |
| additional_images | No | Up to 2 additional reference image URLs | |
| aspect_ratio | No | Output aspect ratio (default: match input image) | match_input_image |
| seed | No | Random seed for reproducible results |
Implementation Reference
- mcp/vap_mcp_proxy.py:191-218 (handler)The `handle_tools_call` function acts as the central router for tool execution in the VAP MCP Proxy. While it currently explicitly handles `generate_video`, `estimate_video_cost`, and `get_task`, other tools (including potential future 'ai_edit' tools) are forwarded directly to the backend API via `make_request`.
def handle_tools_call(params: Dict) -> Dict: """ Handle tools/call request. Directive #240: Special handlers for video tools. """ tool_name = params.get("name", "") arguments = params.get("arguments", {}) # ═══════════════════════════════════════════════════════════════════ # VIDEO TOOL HANDLERS (Directive #240) # ═══════════════════════════════════════════════════════════════════ if tool_name == "generate_video": return _handle_generate_video(arguments) if tool_name == "estimate_video_cost": return _handle_estimate_video_cost(arguments) if tool_name == "get_task": return _handle_get_task(arguments) # Default: forward to MCP API response = make_request("/tools/call", { "name": tool_name, "arguments": arguments }) return response