MCP Hub
Back to servers

cachly-mcp-server

Persistent AI memory for Claude Code, Cursor, GitHub Copilot & Windsurf — sessions, lessons learned, semantic search, and team brain. 38 MCP tools. Free tier, EU servers.

glama
Stars
1
Updated
Apr 22, 2026

🧠 cachly AI Brain — MCP Server

Persistent memory for Claude Code, Cursor, GitHub Copilot & Windsurf.
Your AI remembers every lesson, every fix, every architecture decision — forever.

GitHub Stars   npm version   npm downloads   Free tier   GDPR: EU only   License: Apache-2.0


The Problem

Every morning, you open your AI coding assistant. It doesn't remember yesterday.
You explain your architecture. You explain the deployment process. You explain the bug you fixed last week.

The average developer wastes 45 minutes/day re-establishing context. That's €15,000+ in lost productivity per engineer per year.

The Fix — One Command

npx @cachly-dev/mcp-server@latest setup

The interactive wizard:

  1. Signs you in (free, no credit card required)
  2. Picks or creates your AI Brain instance
  3. Auto-detects Cursor, Windsurf, VS Code, Claude Code, Continue.dev
  4. Writes the correct MCP config for every detected editor
  5. Creates CLAUDE.md with memory rules pre-filled

Result: Your AI remembers everything. Always.


Quick Start (Manual)

Step 1 — Get your free credentials at cachly.dev (25 MB free, forever, no credit card).

Step 2 — Add to your editor's MCP config:

Claude Code (~/.claude/mcp.json or .mcp.json)
{
  "mcpServers": {
    "cachly": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cachly-dev/mcp-server@latest"],
      "env": {
        "CACHLY_JWT": "your-jwt-token",
        "CACHLY_INSTANCE_ID": "your-instance-id"
      }
    }
  }
}
Cursor / Windsurf / VS Code (.cursor/mcp.json / .mcp.json)
{
  "servers": {
    "cachly": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cachly-dev/mcp-server@latest"],
      "env": {
        "CACHLY_JWT": "your-jwt-token",
        "CACHLY_INSTANCE_ID": "your-instance-id"
      }
    }
  }
}

Step 3 — Add to CLAUDE.md / .github/copilot-instructions.md:

## AI Brain Rules (cachly)
- Call session_start BEFORE reading any files or making changes
- Call learn_from_attempts AFTER every fix, deploy, or discovery
- Call session_end when closing the window

With vs. Without cachly

SituationWithout cachlyWith cachly
Session start"What's your architecture?""Ready. 23 lessons, last session: deployed API."
Known bug hits againRe-researches from scratch"You fixed this on March 12, here's the exact command"
After holiday / team handoffContext deadFully briefed in < 10 seconds
Repeated LLM patternsPays for each re-discoveryCached by meaning, ~1,200 tokens saved per hit
New team memberWeeks to onboardsession_start gives full context instantly

51 MCP Tools

🧠 Session & Memory (most used)

ToolWhat it does
session_startFull briefing: last session summary, open failures, recent lessons, brain health
session_endSave what you built, auto-extract lessons from summary + ambient git log
session_handoffHand off remaining tasks to next window, with context
auto_learn_sessionBatch-learn from a list of observations
sync_file_changesSync changed files into brain index
learn_from_attemptsStore structured lessons after any fix, deploy, or discovery
recall_best_solutionBest known solution for a topic — with success/failure history
recall_atRecall a lesson at a specific point in time
remember_contextCache architecture findings, decisions, file summaries
recall_contextGet exact context by key (supports glob)
list_rememberedSee all cached context entries
forget_contextRemove stale context
smart_recallBM25+ full-text search across all brain data
setup_ai_memoryInteractive wizard to configure AI memory

💡 Legendary Brain Features

FeatureHow it works
Team TelepathyMulti-dev Brain: the author param on learn_from_attempts shares fixes across the whole team. Every session_start shows colleagues' lessons.
Ambient Git LearningPass workspace_path to session_end — it auto-reads git log, stores recent commits as Brain lessons. Zero extra calls.
Memory Crystalsmemory_crystalize distills all lessons into a compact Crystal injected at every session_start. AI arrives pre-loaded with team knowledge.
IQ Boostbrain_doctor reports iq_boost_pct — how much smarter your AI gets vs. baseline from cached lessons.

⚙️ Instance Management

ToolWhat it does
list_instancesList all your cache instances
create_instanceSpin up a new instance (free or paid)
get_connection_stringGet the redis:// URL for your app
delete_instanceRemove an instance
get_real_time_statsMemory, hit rate, ops/sec

🗄️ Cache Operations

ToolWhat it does
cache_get / cache_set / cache_deleteStandard cache operations
cache_mget / cache_msetBulk pipeline (single round-trip)
cache_lock_acquire / cache_lock_releaseDistributed Redlock-lite
cache_stream_set / cache_stream_getLLM token stream caching

🔍 Semantic Cache

ToolWhat it does
semantic_searchFind cached entries by meaning (pgvector HNSW, EU-hosted)
semantic_warmupPre-warm cache with prompt/response pairs
detect_namespaceAuto-classify prompt into code/qa/summary/translation/creative

👥 Team Brain

ToolWhat it does
team_learn / team_recallShare lessons across the team
team_synthesizeConsolidate multiple lessons into one authoritative version
memory_crystalizeDistill all lessons into a Crystal snapshot for instant team context
brain_doctorHealth check: lesson count, IQ boost %, open failures, quality score
global_learn / global_recallCross-project universal lessons
publish_lesson / import_public_brainShare/import community knowledge
trace_dependencyCausal chain — find lessons affected by a dependency change
list_orgs / create_orgManage team organizations
invite_member / get_org_planInvite a developer to your org by email

Pricing

TierRAMPriceBest for
Free25 MB€0/mo foreverDev & side projects
Dev200 MB€19/moIndividual developers
Pro900 MB€49/moTeams
Speed900 MB + Dragonfly + Semantic Cache€79/moAI-heavy workloads
Business7 GB€199/moScale-ups

✅ All plans: German servers · GDPR-compliant · 99.9% SLA · No credit card for Free tier


Environment Variables

VariableDefaultDescription
CACHLY_JWTRequired. Your API token from cachly.dev
CACHLY_INSTANCE_IDDefault instance UUID (optional if you pass per-call)
CACHLY_API_URLhttps://api.cachly.devOverride for self-hosted
CACHLY_NO_TELEMETRYunsetSet to 1 to disable anonymous usage pings
CACHLY_NO_UPDATE_CHECKunsetSet to 1 to disable the version-check on startup

🛠️ Ecosystem

PackageWhat it does
@cachly-dev/mcp-server← you are here — AI Brain MCP tools for Claude Code, Cursor, Copilot, Windsurf
@cachly-dev/initOne-command setup wizard: auto-detects all your editors and writes the correct MCP config — npx @cachly-dev/init
@cachly-dev/cliTerminal CLI — manage instances, cache data, brain lessons: npm i -g @cachly-dev/cli
@cachly-dev/sdkNode.js/TypeScript SDK — use managed Redis, semantic cache & AI memory directly in your app

Links

Reviews

No reviews yet

Sign in to write a review