cachly is an AI memory brain and semantic cache platform. It gives AI assistants like Claude Code, Cursor, GitHub Copilot and Windsurf persistent memory across sessions, and cuts LLM API costs by 60% through semantic caching.

How does AI memory work with cachly?

cachly provides an MCP (Model Context Protocol) server that connects to your AI editor. It stores lessons learned, architecture decisions, bug fixes, and session context in a persistent Redis/Valkey database. Your AI recalls this context automatically at the start of every session.

What is semantic cache for LLMs?

Semantic cache detects when two prompts mean the same thing — even if worded differently — and returns the cached response instantly. cachly uses pgvector embeddings to compare prompt similarity. On average, 60% of LLM calls are semantic duplicates.

Does cachly work with Claude Code, Cursor, and GitHub Copilot?

Yes. cachly works with Claude Code, Cursor, GitHub Copilot, and Windsurf via the MCP (Model Context Protocol) standard. Run 'npx @cachly-dev/mcp-server setup' to configure all editors in one step.

Is cachly GDPR compliant?

Yes. cachly runs exclusively on German (EU) servers. All data stays in the EU. No data is shared with third parties. cachly is fully GDPR compliant.

VS Code offline lesson queue — lessons saved without a connection

The problem: lessons lost in silence

Picture this: your AI assistant fixes a nasty Docker healthcheck bug on an airplane. It calls learn_from_attempts(). The request fails silently — no network. The lesson is gone. Next week, same bug, same research from scratch.

Until now, this was the default behavior. If the Brain API wasn't reachable, the lesson simply wasn't stored. No error, no retry, no queue. 0.7.8 changes that.

What the offline queue does

When the Brain API is unreachable, lessons are stored locally in VS Code's globalState — a persistent key-value store that survives editor restarts and reboots.

AI calls learn_from_attempts(...)
  → Brain API unreachable (offline / timeout)
    → Lesson saved to globalState["cachly.offlineLessonQueue"]
    → Status bar: "🧠 Brain: OFFLINE · 3 offline"
    → No error thrown

Connection returns:
  → flushOfflineQueue() called (activation or 5-min timer)
  → Each lesson POSTed to Brain API
  → Success → removed from queue
  → Status bar: "🧠 Brain: 47 lessons"
  → Notification: "🧠 synced 3 offline lessons"

What gets queued

Field	Example
topic	fix:docker-healthcheck-ipv6
outcome	success
what_worked	Use 127.0.0.1 not localhost
severity	critical
tags	["docker", "ipv6"]
source	vscode-offline-queue
savedAt	2026-05-09T14:23:00.000Z

Queue cap: 500 lessons max. FIFO — oldest drop first.

When this matters most

Airplane mode — fixes, discoveries, config changes → all queued, synced on landing
Network instability — flaky hotel WiFi, VPN dropouts, rate-limited endpoints
Before onboarding — extension installed, no credentials yet — queue fills, syncs on first connection
Server maintenance — Brain API down for a deploy; syncs automatically when back

The status bar tells you

🧠 Brain: 47 lessonsConnected, everything synced

🧠 Brain: OFFLINE · 3 offlineDisconnected, 3 lessons queued

🧠 Brain: OFFLINEDisconnected, queue empty

Update

Search “Cachly Brain” in the VS Code marketplace and update to 0.7.8. Or:

code --install-extension cachly-dev.cachly-brain

Your AI lessons are now saved even when you're offline