Roadmap

This page describes the current main branch and future directions, without committing to release dates or priorities. For packaged versions, check release notes.

Implemented on main

• Context and retrieval: L0/L1/L2 context layers, Viking URIs, semantic search, and context-aware retrieval. Concepts
• Resources: document, code, web, and media imports; scheduled refreshes for re-readable sources. Audio/video files can be stored; understanding requires an enabled, compatible VLM. Resources
• Updates and history: freshness-aware parent-summary refreshes and Git-backed snapshot commit, history, and restore. Parent refreshes can be deferred; snapshots require explicit commits and restore file content, not historical ACLs or vector indexes. Context layers · Snapshots
• Sessions and memory: conversation tracking, memory extraction, and session archival. Sessions
• Access and integration: HTTP API, SDKs, CLI, MCP, and agent plugins. API overview · MCP guide
• Operations: JSON configuration (ov.conf), multiple model providers, tenant isolation, encryption, observability, and local/S3 storage. Configuration · Deployment

Future directions

• Further distributed-storage development.
• More agent-framework adapters.

Track proposals and discuss scope in GitHub issues. To contribute, see the contribution guide.