Portable, cross-vendor AI memory in .fafm — offline-first SDK
-
Updated
Aug 15, 2026 - Python
#
fafm
Here are 6 public repositories matching this topic...
FAF Memory plugin for Claude Code — Permanent Memory Layer (PML). Wraps faf-memory-mcp via uvx. .fafm — IANA-registered cross-vendor persistent AI memory. 412× faster type-filter vs grep (falsifiable receipt: Wolfe-Jam/faf-memory-proof).
-
Updated
Jul 21, 2026
Grok + Claude co-built memory receipts: .fafm / .fafb structured tier delivers 436× to 1,399× type-filter speedup vs grep, measured on Smithsonian Open Access + Claude memory corpora. Methodology mirrors faf-memory-proof's 412×.
-
Updated
May 27, 2026 - Python
Falsifiable receipt for the .fafm binary tier — 412× faster type-filter queries vs grep on a 492-file AI memory corpus. Methodology + scripts + sanitized pilot. Built same day as the IANA vnd.fafm+yaml registration.
-
Updated
May 27, 2026 - Python
MCP server for .fafm — Permanent Memory Layer (PML). Wraps claude-fafm-sdk via fastmcp. IANA-registered cross-vendor persistent memory. 412× faster type-filter queries vs grep (see faf-memory-proof for falsifiable receipt).
-
Updated
Aug 15, 2026 - Python
Python SDK that gives xAI Grok Voice agents persistent memory across sessions, devices, and models. LiveKit-enabled. MIT licensed.
python ai mcp realtime developer-tools persistent-memory vml agents grok xai faf voice-ai voice-agent livekit llm model-context-protocol ai-context paralinguistic voice-memory-layer fafm
-
Updated
Aug 15, 2026 - Python
Improve this page
Add a description, image, and links to the fafm topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the fafm topic, visit your repo's landing page and select "manage topics."