Skip to content

Roadmap

Roadmap

Rightsize AI context

  • Deterministic orchestration (conditional steps, for-each steps, parallel-join)
  • Agentic with A2A server adapter with tool discovery and execution
  • Support dynamic MCP tools search
  • Add ikanos import mcp to inspect any existing MCP server and generate an exposes: type: mcp capability to curate down from
  • Facilitate skills publication in skills marketplaces
  • Facilitate integration with MCP and AI gateways
  • MCP trust propagation
  • Enable interactive MCP Apps
  • Evolve MCP server adapter to support server-side code mode like Cloudflare
  • Add AI framework wrappers (Langchain4j, LangChain, LlamaIndex)
  • Support AI Catalog spec (join MCP, A2A initiative)

Reuse existing APIs & Data

  • Add support for FILE and SQL as client adapters
  • Add support for gRPC consumption (incl. proto import)
  • Increase HTTP client resiliency
  • Retry on error or timeout
  • Time limiter
  • Circuit breaker
  • Rate limiter
  • Bulkhead
  • Cache
  • Fallback
  • Support Webhook server adapter for workflow automation
  • Facilitate integration with API gateways
  • Enable token refresh flows for consumed APIs
  • CORS configuration for API developer portals
  • Gateway context propagation via Open Telemetry
  • mTLS client certificates on consumes
  • Support HTTP cache control directives
  • Enable pagination at consumes and exposes level
  • Enhance mocking feature for APIs either exposed or consumed
  • Add client SDKs generation to Ikanos CLI for top languages (TypeScript, Python, Java, Go)
  • Ensure it is extensible to bring your own client SDK generator (Apimatic, Fern, Stainless, Speakeasy, etc.)
  • Add support for Singer, Airbyte and SQL as server adapters

Core developer experience

  • Native integration with Langchain4j, see issue #293
  • Publish starter capability templates (golden path skeletons with all required fields pre-filled)
  • Publish Naftiko JSON Structure
  • Add live and generated capability UI similar to Swagger UI, Redocly and others
  • Publish reference bridge capabilities (RSS/Atom XML feeds, XML/SOAP, CSV, etc.)
  • Solidify the existing beta version scope
  • Increase test coverage and overall quality
  • Publish JSON Schema to JSON Schema Store
  • Profile and optimize engine for best latency and throughput
  • Provide Control webapp (per Capability)
  • Publish Docker Desktop Extension to Docker Hub