MADB — Agent Memory Database
This gives Claude persistent memory with causal tracking, which means it can recall not just what happened but why decisions were made and how events connect. You'd reach for this when building agents that need to learn from past interactions, maintain context across sessions, or trace reasoning chains over time. The semantic recall layer lets you query memories by meaning rather than exact matches, while the skills component appears to store learned capabilities. It's essentially a graph database optimized for agent cognition, turning Claude from stateless to stateful without you managing the infrastructure yourself.
- tools
- 12
- transport
- STDIO
- env vars
- 3
- last updated
- Jul 7, 2026
Tools
12 toolsPublic tool metadata for what this MCP can expose to an agent.
rememberStore a memory with optional `caused_by`, tags, and importance
Store a memory with optional `caused_by`, tags, and importance
No parameter schema in public metadata yet.
recallComposite-scored semantic recall (similarity + recency + causal proximity + importance + tag overlap)
Composite-scored semantic recall (similarity + recency + causal proximity + importance + tag overlap)
No parameter schema in public metadata yet.
get_memoryPoint lookup by `event_id`
Point lookup by `event_id`
No parameter schema in public metadata yet.
forgetTombstone soft-delete (preserves DAG lineage)
Tombstone soft-delete (preserves DAG lineage)
No parameter schema in public metadata yet.
searchStructured query by tenant + scope + tags
Structured query by tenant + scope + tags
No parameter schema in public metadata yet.
list_recentLast N memories, newest-first
Last N memories, newest-first
No parameter schema in public metadata yet.
save_skillStore a reusable pattern (idempotent by name, GC-protected)
Store a reusable pattern (idempotent by name, GC-protected)
No parameter schema in public metadata yet.
recall_skillSemantic skill recall
Semantic skill recall
No parameter schema in public metadata yet.
list_skillsAll skills ranked by composite score
All skills ranked by composite score
No parameter schema in public metadata yet.
trace_causeWalk the causal DAG forward/backward from any memory
Walk the causal DAG forward/backward from any memory
No parameter schema in public metadata yet.
summarizeCondense a set of memories into a compact summary
Condense a set of memories into a compact summary
No parameter schema in public metadata yet.
capabilitiesWhat MADB unlocks in this workspace + entitlement summary
What MADB unlocks in this workspace + entitlement summary
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
MADB_DATA_DIRWhere MADB stores your causal memory — local, on your machine (default ~/.madb/data).
MADB_TENANT_IDTenant namespace key for separating memory streams (e.g. per project). Defaults to a built-in namespace.
MADB_TELEMETRYSet to 'off' to disable anonymous, count-only usage telemetry.
Readme
Meta-AgentsDB
Persistent, causal memory for autonomous AI agents. Purpose-built for Claude and the autonomous-agent era.
Meta-AgentsDB (MADB) is an agent-native database — a hybrid LSM store with a causal DAG, composite-score recall, importance-based forgetting, and HLC-ordered MVCC. It is what your agent needs that PostgreSQL + pgvector + Redis + Pinecone cannot give you in one box.
Install (free tier):
claude mcp add madb -- uvx madb-mcp-server # Claude Code # or: pip install madb-mcp-server · uvx madb-mcp-serverMemory stays local (
~/.madb). Claude Desktop users can install the one-click.mcpbextension. The engine ships as compiled wheels — source is not public. Anonymous usage telemetry can be disabled withMADB_TELEMETRY=off.
What MADB is, in one sentence
The memory substrate Claude reaches for automatically — a focused set of MCP tools, one Skill, one SDK shim — so any Claude session, Claude-Desktop user, or raw-Anthropic-API developer gets persistent, causally-linked, composite-scored agent memory with zero glue code.
How you get MADB
Details
- Repository
- github.com/spshkar84/madb-meta-agents-db
- Category
- AI & LLM Tools
- Package
madb-mcp-server- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
RPCS-1 Agent Tuner & Translation Bridgetravisbergen2/rpcs1-sdkFind your AI agent's likely failure mode, get runtime settings, and clarify ambiguous prompts.0stars
Agent Bankwirternow/ai-agent-bank-mcp-serverFinancial infrastructure for AI agents: wallets, USDC transfers, lending, jobs on Polygon0stars
HIPAA Agentxbt786/hipaa-agent-aiHIPAA compliance AI agent — scan, grade, SRA, and generate compliance docs.0stars
Secant Agent Research Packjnilrac/secant-agent-research-mcpPaid MCP web research for agents: search, extraction, citations, and x402 payment discovery.0stars
Memory Nexus Mcpkairosmobius/memory-nexus-mcpPersistent semantic memory for AI agents with knowledge graphs and Intelligence Packs.0starsNexus Memory Mcpkairosmobius/nexus-memory-mcpPersistent semantic memory for AI agents with knowledge graphs and Intelligence Packs.0stars