Mem0 MCP Serverofficial
A
licenseA
qualityF
maintenanceAn MCP server that integrates with mem0.ai to help users store, retrieve, and search coding preferences for more consistent programming practices.
2
9
658
Apache 2.0
dakera-mcp
Officialby Dakera-AI
The dakera-mcp server provides AI agents with a persistent, queryable memory layer featuring smart token management. Here's what you can do:
Store and Manage Memories
Persist facts, decisions, or context with importance weighting (0.0–1.0), tags, memory type (episodic/semantic/procedural/working), optional expiry, and session association (dakera_store).
Recall and Search Memories
Semantic recall (
dakera_recall) — Retrieve top-k memories by vector similarity, with optional time filters and knowledge-graph expansionAdvanced search (
dakera_search) — Semantic search pre-filtered by tags and/or memory typeFull-text search (
dakera_fulltext_search) — BM25 keyword search for exact-term recall (error codes, IDs, names)Hybrid search (
dakera_hybrid_search) — Combined BM25 + vector search in a single pass, tunable viavector_weightBatch recall (
dakera_batch_recall) — Filter-based listing by tags, importance range, time window, type, or session
Delete Memories
dakera_forget— Delete specific memories by ID or bulk-delete by tagdakera_batch_forget— Bulk-delete by filter criteria (tags, importance range, time window, memory type)
Session Management
dakera_session_start— Open a session to group related memories with optional metadatadakera_session_end— Close a session with an optional summary
Knowledge Graph & Extraction
dakera_knowledge_graph— Build a knowledge graph from a seed memory by exploring connections via embedding similaritydakera_extract— Extract structured entities, topics, key phrases, and summaries from free-form text using configurable providers (GLiNER, OpenAI, Anthropic, Ollama, etc.)
Dynamic Tool Discovery & Loading
dakera_discover_tools— Search the tool catalog by keyword or tier without loading full schemasdakera_load_tools— Fetch full input schemas for specific tools on demand to minimize token usage
Key Highlights
Default 14-tool core profile uses ~2,964 tokens; expand to 86 tools via profiles (
admin,power,all) or on-demand loadingScored 88.2% on the LoCoMo benchmark (1,540 questions)
Compatible with Claude, Claude Code, and any MCP-compatible framework
Self-hostable via Docker or Kubernetes
⚡ dakera-mcp
MCP server for Dakera AI. Gives any MCP-compatible AI agent persistent, queryable memory — with smart token management built in.
Works with Claude, Claude Code, and any MCP-compatible framework.
Part of Dakera AI — the memory engine for AI agents.
The Dakera memory engine scores 88.2% Recall@20 on LoCoMo (1,540 questions · LLM-judge scored) — benchmark details
Architecture: 14 core tools + on-demand discovery
Starting every agent session with 60+ tool schemas wastes ~15K tokens before you write a single message. dakera-mcp solves this with hybrid tool exposure:
14 tools loaded by default — the 12 highest-frequency memory operations + 2 meta-discovery tools
On-demand expansion — use
dakera_discover_toolsanddakera_load_toolsto fetch additional tool schemas only when you need them
Default tool set (core profile)
Tool | Purpose |
| Store a memory with importance, tags, and type |
| Semantic recall by query text |
| Advanced memory search with tag/type filters |
| Start a session to group related memories |
| End a session with optional summary |
| Bulk filter-based recall (by tags, importance, time) |
| Delete specific memories by ID |
| Combined vector + BM25 search |
| BM25 full-text search |
| Build a knowledge graph from a seed memory |
| Extract entities and structure from free-form text |
| Bulk delete by tags, type, or time range |
| Search the full tool catalog by keyword or tier |
| Load full schemas for specific tools on demand |
Profiles & token cost
Profile | Tools | ~Tokens | How to enable |
core | 14 | ~2,964 | Default — always loaded |
admin | 32 | ~5,975 |
|
power | 69 | ~13,205 |
|
all | 87 | ~16,212 |
|
Accessing additional tools
# In your agent: discover what's available
dakera_discover_tools(tier="power")
→ returns names + descriptions, no schemas loaded
# Load schemas for the tools you want
dakera_load_tools(tools=["dakera_consolidate", "dakera_agent_stats"])
→ returns full inputSchema for each toolProfile selection
The profile controls which tools appear in tools/list. Three ways to set it:
1. Per-request (in tools/list params):
{"profile": "power"}2. Environment variable (applies to all requests):
DAKERA_MCP_PROFILE=power3. Default: core (14 tools, ~2,964 tokens)
Related MCP server: Zep MCP Server
Run Dakera
The MCP server connects to a Dakera memory server. You need one running first:
docker run -d \
--name dakera \
-p 3300:3000 \
-e DAKERA_ROOT_API_KEY=dk-mykey \
ghcr.io/dakera-ai/dakera:latestFor persistent storage (recommended):
curl -sSfL https://raw.githubusercontent.com/Dakera-AI/dakera-deploy/main/docker-compose.yml \
-o docker-compose.yml
DAKERA_API_KEY=dk-mykey docker compose up -d
curl http://localhost:3000/health # → {"status":"ok"}Full deployment guide (Docker Compose, Kubernetes, Helm): dakera-deploy
Install
npm / npx (Node.js 18+)
# Global install
npm install -g @dakera-ai/dakera-mcp
# Or run directly without installing
npx @dakera-ai/dakera-mcpHomebrew (macOS / Linux)
brew install dakera-ai/tap/dakera-mcpCargo
cargo install dakera-mcpDocker
docker pull ghcr.io/dakera-ai/dakera-mcp:latestBinary download
Pre-built binaries for macOS, Linux, and Windows are available on the releases page.
Platform | File |
macOS (Apple Silicon) |
|
macOS (Intel) |
|
Linux x64 |
|
Linux arm64 |
|
Windows x64 |
|
Connect
Add to .mcp.json (Claude Code) or claude_desktop_config.json (Claude Desktop):
{
"mcpServers": {
"dakera": {
"command": "dakera-mcp",
"env": {
"DAKERA_API_URL": "http://localhost:3300",
"DAKERA_API_KEY": "your-key"
}
}
}
}To start with the power profile (exposes 68 tools):
{
"mcpServers": {
"dakera": {
"command": "dakera-mcp",
"env": {
"DAKERA_API_URL": "http://localhost:3300",
"DAKERA_API_KEY": "your-key",
"DAKERA_MCP_PROFILE": "power"
}
}
}
}Why This Exists
AI agents forget everything when the session ends. Dakera fixes that. This MCP server gives your agent a persistent memory layer with zero infrastructure overhead — point it at a Dakera instance and it works.
The 14-tool default keeps your context window lean. The meta-tools let you expand on demand when you need advanced operations like bulk vector upsert, knowledge graph traversal, or memory federation.
→ dakera.ai for hosted instance
→ Self-host with dakera-deploy
Documentation
Related
Repo | What it is |
Python SDK | |
TypeScript SDK | |
CLI | |
Self-host Dakera |
dakera.ai · Documentation · Request Early Access
Part of the Dakera AI open-core ecosystem. Built with Rust. Self-hosted. Zero dependencies.
Available Tools
1 tooldakera_hybrid_searchA
BM25 + vector ANN hybrid search in a single pass. Omit vector for BM25-only mode. Use for RAG when pure semantic or keyword search alone is insufficient. vector_weight: 0.0=BM25, 1.0=vector (default 0.5).
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes | Text query for full-text search | |
| top_k | No | Number of results to return | |
| filter | No | Optional metadata filter | |
| vector | No | Query embedding; omit for BM25-only. | |
| namespace | Yes | Namespace to search in | |
| vector_weight | No | Vector score weight 0.0–1.0; text weight = 1−value. | |
| include_vectors | No | ||
| include_metadata | No |
TDQS
A4.1/5.0
Behavior3/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description explains the hybrid algorithm and weight behavior, but does not disclose safety, permissions, cost, or error behavior. The 'single pass' note gives a performance hint, but overall transparency is limited.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise, front-loaded sentences with no redundancy. Every sentence provides essential information, and the structure is clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness3/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 8 parameters, nested filter object, and no output schema, the description is somewhat minimal. It does not explain return values, pagination, or results handling, leaving gaps for a complex search tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters4/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 75% (two parameters lack descriptions in schema). The description adds meaning by explaining the vector parameter for BM25-only mode and the vector_weight range, enhancing understanding beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose5/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly specifies a hybrid search combining BM25 and vector ANN in a single pass, and distinguishes from siblings by noting BM25-only mode and use case for RAG when pure methods are insufficient.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines4/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly states when to use ('RAG when pure semantic or keyword search alone is insufficient') and how to configure modes (omit vector for BM25-only, adjust vector_weight). Lacks explicit exclusion criteria but provides clear context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
13 tool updates
v0.10.11- Removed
dakera_batch_forget - Removed
dakera_batch_recall - Removed
dakera_discover_tools - Removed
dakera_extract - Removed
dakera_forget - Removed
dakera_fulltext_search - Removed
dakera_knowledge_graph - Removed
dakera_load_tools - Removed
dakera_recall - Removed
dakera_search - Removed
dakera_session_end - Removed
dakera_session_start - Removed
dakera_store
14 tool updates
v0.10.8- Added
dakera_batch_forget - Added
dakera_batch_recall - Added
dakera_discover_tools - Added
dakera_extract - Added
dakera_forget - Added
dakera_fulltext_search - Added
dakera_hybrid_search - Added
dakera_knowledge_graph - Added
dakera_load_tools - Added
dakera_recall - Added
dakera_search - Added
dakera_session_end - Added
dakera_session_start - Added
dakera_store
TDQS
A4/5.0
Disambiguation5/5
With only one tool, there is no risk of confusion between tools, so disambiguation is perfect.
Naming Consistency5/5
A single tool has no naming inconsistencies, so this dimension is trivial.
Tool Count3/5
One tool is minimal; while it covers a core function, the server feels thin for a typical MCP server scope.
Completeness2/5
Only a search tool is provided; missing common operations like document retrieval, indexing, or configuration, making the surface incomplete for a search server.
Maintenance
ActivitySlowing
ResponsivenessUnresponsive
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Cloud-hosted MCP server for durable AI memory
- memnodeOAuthdev.memnode
Persistent, inspectable memory for AI agents with lineage, correction, and a hosted MCP endpoint.
Persistent memory for AI agents across Claude, ChatGPT and any MCP client.
Cloudflare Workers MCP server: agent-memory
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables Claude to store, search, and retrieve persistent memories using Zep Cloud's thread-based memory system for maintaining context across conversations.-
- AlicenseNot gradedqualityAmaintenanceA graph-based MCP server that provides AI coding agents with persistent memory to store patterns, track complex relationships, and retrieve knowledge across sessions. It leverages graph structures to handle temporal queries and relational paths that traditional vector stores often miss.244MIT
Recallofficial
AlicenseNot gradedqualityCmaintenanceOpen-source MCP memory server for AI agents — persistent, searchable, tiered memory across sessions. Works over stdio (Cursor, Claude Desktop) or HTTP+SSE. MIT licensed.7MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Dakera-AI/dakera-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server