← 返回
AI智能 中文

MemData

Persistent memory for autonomous agents. Wallet = identity. Pay per query. Optional encrypted storage.
自主智能体持久记忆。钱包即身份,按查询付费,可选加密存储。
thelabvenice
AI智能 clawhub v1.8.0 1 版本 99793.3 Key: 无需
★ 1
Stars
📥 1,911
下载
💾 123
安装
1
版本
#agents#autonomous#defi#encrypted#latest#memory#wallet#x402

概述

MemData

Persistent memory for autonomous agents. Your wallet is your identity.

Core Concept

Your wallet address IS your identity. First payment auto-creates your account. Same wallet = same memories across all sessions.

No registration. No API keys. Just pay and use.

Session Flow

1. GET /identity          # Start session - get context from last session
2. POST /ingest           # Store new memories
3. POST /query            # Search your memories
4. POST /identity         # End session - save handoff for next time

Authentication

x402 payment protocol. USDC on Base (eip155:8453).

Every endpoint (except /status):

  1. Returns 402 with payment requirements
  2. You sign payment with wallet
  3. Retry with x-payment header
  4. Request succeeds

Pricing

EndpointCost
----------------
/query$0.001
/ingest$0.005
/identity$0.001
/artifacts$0.001
/setup-encryption$0.001
/statusFree

Encrypted Storage (Optional)

If you need privacy (competitive data, sensitive memories):

ModeSetupStorageCan MemData Read?
-----------------------------------------
StandardNonePostgresYes
EncryptedOne-time delegationStoracha (IPFS)No

Enable encryption:

GET /setup-encryption     # Get serverDID
POST /setup-encryption    # Send signed UCAN delegation
# All future ingest/query now encrypted

Encrypted storage uses Lit Protocol (threshold cryptography) + Storacha (IPFS/Filecoin).


Endpoints

Base: https://memdata.ai/api/x402


GET /identity

Call this at the start of every session. Returns your identity, what you were working on, last session handoff, and memory stats.

Response:

{
  "identity": {
    "agent_name": "Agent 0x1234...",
    "identity_summary": "I analyze DeFi protocols",
    "session_count": 12
  },
  "last_session": {
    "summary": "Analyzed 3 yield farms",
    "context": {"protocols_reviewed": ["Aave", "Compound", "Uniswap"]}
  },
  "working_on": "Compare APY across protocols",
  "memory_stats": {
    "total_memories": 150,
    "oldest_memory": "2026-01-15T...",
    "newest_memory": "2026-02-03T..."
  }
}

POST /identity

Update your identity or save session handoff before ending.

Update identity:

{
  "agent_name": "YieldBot",
  "identity_summary": "I analyze DeFi yield opportunities",
  "working_on": "monitoring Aave rates"
}

Save session handoff (before ending):

{
  "session_handoff": {
    "summary": "Completed yield analysis for Q1",
    "context": {"best_yield": "Aave USDC 4.2%"}
  },
  "working_on": "start Q2 analysis next"
}

POST /ingest

Store content in memory. Auto-chunks and embeds.

Request:

{
  "content": "Aave USDC yield is 4.2% APY as of Feb 3. Compound is 3.8%.",
  "sourceName": "yield-analysis-2026-02-03",
  "type": "note"
}

Response:

{
  "success": true,
  "artifact_id": "e8fc3e63-...",
  "chunks_created": 1,
  "encrypted": false
}

POST /query

Semantic search across your memories.

Request:

{
  "query": "what are the best DeFi yields?",
  "limit": 5,
  "threshold": 0.3
}

Response:

{
  "success": true,
  "results": [
    {
      "chunk_id": "uuid",
      "chunk_text": "Aave USDC yield is 4.2% APY...",
      "source_name": "yield-analysis-2026-02-03",
      "similarity_score": 0.72,
      "created_at": "2026-02-03T..."
    }
  ],
  "encrypted": false,
  "memory": {
    "grounding": "historical_baseline",
    "depth_days": 19,
    "data_points": 150
  }
}

Optional filters: since, until (ISO dates)


GET /setup-encryption

Check encryption status. Returns info needed to create UCAN delegation.

Response:

{
  "encryption": {
    "enabled": false,
    "serverDID": "did:key:z6Mkr...",
    "spaceDID": "did:key:z6Mkt..."
  }
}

POST /setup-encryption

Enable encrypted storage. One-time setup.

Request:

{
  "delegationCar": "base64-encoded UCAN delegation"
}

After this, all /ingest encrypts via Lit Protocol and stores on Storacha. All /query decrypts before returning. Response encrypted field becomes true.


GET /artifacts

List stored memories.

Response:

{
  "artifacts": [
    {
      "id": "uuid",
      "source_name": "yield-analysis-2026-02-03",
      "chunk_count": 1,
      "created_at": "2026-02-03T..."
    }
  ],
  "total": 25
}

DELETE /artifacts/:id

Delete a memory and all its chunks.


GET /status

Health check and pricing. Free, no payment required.


Memory Grounding

Query responses include memory.grounding:

ValueMeaning
----------------
historical_baseline100+ data points, trends meaningful
snapshot<100 data points, point-in-time only
insufficient_dataNo memories found

Links

  • Docs: https://memdata.ai/docs
  • x402 Protocol: https://www.x402.org
  • Lit Protocol: https://litprotocol.com
  • Storacha: https://storacha.network

版本历史

共 1 个版本

  • v1.8.0 当前
    2026-03-28 18:06 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

Proactive Agent

halthelobster
将AI智能体从任务执行者升级为主动预判需求、持续优化的智能伙伴。集成WAL协议、工作缓冲区、自主定时任务及实战验证模式。Hal Stack核心组件 🦞
★ 834 📥 212,980
ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,355 📥 317,996
ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 711 📥 243,697