Export, import, and migrate RVF vector memory files for portable agent knowledge transfer.
npx skills add https://github.com/ruvnet/claude-flow --skill rvf-manage
Please help me install this Agent Skill. Skill: rvf-manage Source: https://github.com/ruvnet/claude-flow/tree/main/plugins/ruflo-rvf/skills/rvf-manage Command: npx skills add https://github.com/ruvnet/claude-flow --skill rvf-manage
Agent memory stored in one project or platform is hard to back up, share, or move between environments. This Skill manages RVF (Ruflo Vector Format) files so agent memory—including vector embeddings, metadata, and causal relationships—can be exported, imported, and migrated across projects and teams.
Ask the agent to list all stored memories and export the project-knowledge pattern to an RVF file for transfer to another project.
High-intent search queries and answers about installing and using this skill.
List stored memories with the memory_list tool, then use the hooks transfer tool with the store action to export patterns. You can also run npx @claude-flow/cli hooks transfer store with a pattern flag from the command line.
Export the memory from the source project using the transfer store command, then run hooks transfer from-project with the source path in the target project. This moves RVF-packaged knowledge including embeddings and metadata.
RVF stores 384-dimensional ONNX vector embeddings, metadata such as timestamps, namespaces, and tags, causal relationships between entries, and session context with agent scope.
Yes, the memory_import_claude tool imports existing Claude Code memories into the RVF-backed store, making them available for listing, export, and transfer alongside other entries.
Use the memory_migrate tool, which handles format upgrades for stored entries. After migration, verify the results with memory_stats to confirm storage metrics and entry counts.
Synaptic-Labs-AIStore and retrieve structured memories in a SQLite-backed store for PACT agents.
chicanoandres702Restore project context from memory stores using semantic retrieval and token-budget management.
devs6186Store memory patterns in SQLite with HNSW vector search.
ricableRun embedded vector searches with SQL, SPARQL, and Cypher via WASM.
