-
Notifications
You must be signed in to change notification settings - Fork 1
Roadmap
Jerome Louvel edited this page Apr 12, 2026
·
76 revisions
The goal of this version is to solidify the MVP to enable common AI integration use cases and grow our community.
- Add mocking feature to MCP server adapter similar to REST server adapter
- Add tool annotations (readOnly, destructive, idempotent, openWorld)
- Add support for authentication in the MCP server adapter
- Add HTML and Markdown data format support for HTTP consumption
- Add interoperability with OpenAPI Specification
- Import OAS into an HTTP "consumes" adapter
- Export OAS from a REST "exposes" adapter
- Factorize capability core with functions initially, entities and events later
- Use named objects for input and output parameters, like for properties, matching the JSON Structure syntax
- Enhance traceability and debuggability of engine (support Open Telemetry)
- Support inline scripting steps (JavaScript and Python initially)
- Publish Naftiko JSON Structure
- Publish Naftiko Skill based on Naftiko CLI
- Publish Naftiko Ruleset based on Spectral
- Publish Maven Artifacts to Maven Central
- Publish Javadocs to Javadoc.io
- Publish Docker Image to Docker Hub
- Facilitate integration with MCP and AI gateways
- Facilitate skills publication in skills marketplaces
- Externalize individual "exposes" objects into separate files, similar to "consumes" objects
- Enhance support for tags and labels across capabilities
- Advanced error handling and recovery strategies
- Support HTTP cache control directives
- Enable API token refresh flows
- Enable pagination at consumes and exposes level
- Support Webhook server adapter for workflow automation
- Facilitate integration with API gateways
- Support A2A server adapter with tool discovery and execution
- Allow reuse of "binds" blocks across capabilities
- Add conditional steps, for-each steps, parallel-join
- Publish starter capability templates (golden path skeletons with all required fields pre-filled)
- Expand inline scripting steps (Ruby and Groovy)
- Provide Control port usable via REST clients, Naftiko CLI
- Native integration with Langchain4j, see issue #293
The goal of this version is to deliver a stable MVP, including a stable Naftiko Specification.
- Rightsize AI context
- Evolve MCP server adapter to support server-side code mode like CloudFlare
- Enhance API reusability
- Add support for resiliency patterns (retry, circuit breaker, rate limiter, time limiter, bulkhead, cache, fallback)
- Publish reference bridge capabilities (RSS/Atom XML feeds, XML/SOAP, CSV, etc.)
- Enable capabilities governance
- Expand support for "tags" and "labels" in Naftiko Spec
- Provide enhanced security
- Facilitate authorization management
- Incorporate community feedback
- Complete test coverage and overall quality
The goal of this version is to release our first version ready for production.
- Incorporate community feedback
- 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
The goal of this version is to broaden the platform surface area based on production learnings.
- Add support for gRPC and tRPC as server adapters
- Add full resiliency patterns (rate limiter, time limiter, bulkhead, cache)
- Add support for Singer, Airbyte and SQL as server adapters
- Add support for FILE and SQL as client adapters
- Support templatized SQL request with proper security
- Facilitate integration with Keycloak, OpenFGA
- Provide Control webapp (per Capability)
- Publish Docker Desktop Extension to Docker Hub
- Fabric discovery of published capabilities for consumers