1AIVault1AIVault

1AIVault · Blog

Notes from the vault

Field notes on building one portable memory across Claude, ChatGPT, Cursor, Cline, and Windsurf. MCP patterns, context design, and practical workflow strategies for people using multiple AI tools.

Make AI memory visible — implicit context is the new trust problem

Jun 3, 2026

Make AI memory visible — implicit context is the new trust problem

When AI tools remember things in the background, users feel either magic or unease. Showing exactly what context is carried forward, where it came from, and what got retrieved turns memory from a feeling into a product feature.

trust#trust#memory-visibility#cross-model
AI forgetting is a product-layer problem, not a model flaw

Jun 3, 2026

AI forgetting is a product-layer problem, not a model flaw

Founders writing publicly about "why AI keeps forgetting" are quietly teaching the market a useful frame: memory is something the product layer has to provide. That reframe is where local-first vaults win.

memory#market-education#ai-forgetting#memory-infra
See exactly which memories your AI is actually reading

Jun 1, 2026

See exactly which memories your AI is actually reading

Memory Reads turns raw MCP traffic into a timeline grouped by call, flags empty calls as vault gaps, and surfaces the memories your AI tools reach for most. Plus encrypted vault transfer to a new machine and one-click Claude Code & Codex session resume.

engineering#memory-reads#mcp#claude-code
Stop rebuilding context every time you switch AI tools

May 31, 2026

Stop rebuilding context every time you switch AI tools

Save context once and every AI tool — Claude Desktop, Claude Code, Cursor, Cline, Codex — can read it through a single MCP connection. 1AIVault v1.0.0 ships your portable, local-first AI memory vault.

engineering#ai-memory#mcp#claude-code