AIMLAPI is an AI model access gateway that aggregates 400+ models from different providers behind a single unified API. Chat completions, image generation, embeddings — all accessible through one endpoint. The Naftiko capability for AIMLAPI exposes those model operations as structured MCP tools, making them directly callable from any AI agent or orchestration workflow.
What the Capability Covers
The ai-model-operations capability wraps the AIMLAPI surface and exposes four foundational AI operations as governed tools. Rather than hardcoding model calls into application code, the capability declares them as a specification — typed inputs, structured outputs, authentication handled by the integration layer.
MCP Tools Available
create-chat-completion— Generate a chat response from any of 400+ AI language modelsgenerate-image— Generate images from text prompts using image generation modelscreate-embedding— Create vector embeddings for semantic search and RAG workflowslist-models— Discover all 400+ available models with capabilities and pricing
Capability Specification
naftiko: "1.0.0-alpha1"
info:
label: "AIMLAPI AI Model Operations"
description: "Unified workflow for accessing 400+ AI models via AIMLAPI gateway including chat completions, image generation, embeddings, and model discovery."
tags:
- Artificial Intelligence
- LLM
- Image Generation
- Embeddings
- API Gateway
binds:
- namespace: env
keys:
AIMLAPI_API_KEY: AIMLAPI_API_KEY
capability:
consumes:
- namespace: aimlapi
type: http
baseUri: https://api.aimlapi.com
authentication:
type: bearer
token: ""
resources:
- name: chat
path: "/v1/chat/completions"
operations:
- name: create-chat-completion
method: POST
- name: images
path: "/v1/images/generations"
operations:
- name: create-image
method: POST
- name: embeddings
path: "/v1/embeddings"
operations:
- name: create-embedding
method: POST
- name: models
path: "/v1/models"
operations:
- name: list-models
method: GET
exposes:
- type: mcp
port: 9080
namespace: aimlapi-ai-ops-mcp
transport: http
description: "MCP server for AI-assisted access to AIMLAPI model gateway."
tools:
- name: create-chat-completion
description: "Generate a chat response from any of 400+ AI language models via AIMLAPI"
hints:
readOnly: false
destructive: false
idempotent: false
call: "aimlapi.create-chat-completion"
outputParameters:
- type: object
mapping: "$."
- name: generate-image
description: "Generate an image from a text prompt using AIMLAPI image generation models"
hints:
readOnly: false
destructive: false
idempotent: false
call: "aimlapi.create-image"
outputParameters:
- type: object
mapping: "$."
- name: create-embedding
description: "Generate vector embeddings for semantic search and RAG applications"
hints:
readOnly: true
openWorld: true
call: "aimlapi.create-embedding"
outputParameters:
- type: object
mapping: "$."
- name: list-models
description: "Discover all 400+ available AI models on AIMLAPI platform"
hints:
readOnly: true
openWorld: true
call: "aimlapi.list-models"
outputParameters:
- type: object
mapping: "$."
Running It
docker pull ghcr.io/naftiko/framework:latest
# Set AIMLAPI_API_KEY in your .env file
docker run -p 9080:9080 \
-v ./capabilities/ai-model-operations.yaml:/app/capability.yaml \
--env-file .env \
ghcr.io/naftiko/framework:latest /app/capability.yaml
Once running, any MCP-compatible client can call list-models to discover what is available, then create-chat-completion to invoke completions — all with authentication handled by the capability layer. The AIMLAPI capability is part of the Naftiko Fleet.
