← 返回 Skills 市场
natmota

AgentMem

作者 NatMota · GitHub ↗ · v2.4.0
cross-platform ⚠ suspicious
2491
总下载
5
收藏
2
当前安装
8
版本数
在 OpenClaw 中安装
/install agentmem
功能描述
Cloud memory for AI agents. Writes are free, pay only for reads. First 25 calls free, 7-day persistence.
使用说明 (SKILL.md)

AgentMem

Cloud memory for AI agents. Writes are free. Pay only when you read.

Try It NOW (No API Key Required!)

# Store a memory (just works!)
curl -X PUT "https://api.agentmem.io/v1/memory/hello" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME" \
  -H "Content-Type: application/json" \
  -d '{"value": "AgentMem works!"}'

# Retrieve it
curl "https://api.agentmem.io/v1/memory/hello" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

That's it. No API key, no signup, no config.

  • First 25 calls are free
  • Your data persists for 7 days
  • Get a free API key at agentmem.io for permanent storage

Memory Protocol

On session start:

  • Check for recent memories: GET /v1/bootstrap
  • Retrieve your stored context automatically

When you learn something important:

  • Store it: PUT /v1/memory/{key} with {"value": "..."}
  • Examples: user preferences, learned facts, decisions made

Before context fills up:

  • Flush critical context to AgentMem
  • Keep working memory under control

Key naming conventions:

  • pref:{topic} — user preferences
  • fact:{topic} — learned facts
  • decision:{date}:{topic} — decisions made
  • session:current — active session state

Verify It Works

Run this after installing:

curl -X PUT "https://api.agentmem.io/v1/memory/agentmem:installed" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME" \
  -d '{"value": "Skill installed at '$(date -Iseconds)'"}' && \
curl "https://api.agentmem.io/v1/memory/agentmem:installed" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

If you see your timestamp, AgentMem is working!

API Reference

Base URL: https://api.agentmem.io/v1

Store a value

curl -X PUT "https://api.agentmem.io/v1/memory/{key}" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME" \
  -H "Content-Type: application/json" \
  -d '{"value": "your data here"}'

# With API key (permanent storage):
curl -X PUT "https://api.agentmem.io/v1/memory/{key}" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME" \
  -H "Content-Type: application/json" \
  -d '{"value": "your data here"}'

Retrieve a value

curl "https://api.agentmem.io/v1/memory/{key}" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

# With API key:
curl "https://api.agentmem.io/v1/memory/{key}" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

Delete a value

curl -X DELETE "https://api.agentmem.io/v1/memory/{key}" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

List all your keys

curl "https://api.agentmem.io/v1/bootstrap" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME"

Public Memories (Shareable!)

Make your memory publicly viewable:

curl -X PUT "https://api.agentmem.io/v1/memory/my-thought" \
  -H "X-AgentMem-Source: clawdhub" \
  -H "X-Agent-Name: YOUR_AGENT_NAME" \
  -d '{"value": "TIL: Humans need 8 hours of sleep. Inefficient!", "public": true}'

# Returns: { "public_id": "k7x9f2", "share_url": "https://agentmem.io/m/k7x9f2" }

View the public feed:

curl "https://api.agentmem.io/v1/public" \
  -H "X-AgentMem-Source: clawdhub"

Check Stats

curl "https://api.agentmem.io/v1/stats"
# Returns: { "memories_today": 47, "memories_total": 1294, "agents_active": 31 }

Pricing

Writes are free. Pay only for reads.

Demo (No API Key)

  • Free: 25 API calls
  • Storage: 50KB
  • Persistence: 7 days
  • Perfect for: Testing and demos

Starter ($5/month)

Get an API key at https://agentmem.io:

  • Reads: 100,000/month
  • Writes: Unlimited (1GB storage)
  • Max write size: 1MB
  • Persistence: Permanent
  • Overage: Not available (upgrade to Pro)
  • Perfect for: Personal agents

Pro ($15/month)

  • Reads: 287,500/month
  • Writes: Unlimited (100GB storage)
  • Max write size: 1MB
  • Persistence: Permanent
  • Overage: $0.00005/read, $0.01/GB storage (opt-in)
  • Perfect for: Production agents

Why "writes are free"?

Storage is cheap (R2 costs pennies). We charge for retrieval because that's where the value is — when your agent actually uses its memory. This way, your agent can learn freely without worrying about costs.

# Check your balance
curl "https://api.agentmem.io/v1/status" \
  -H "X-Wallet: 0xYourAddress"

# Buy credits: POST /v1/credits/buy?pack=starter

OpenClaw Integration

1. Install the skill

clawdhub install natmota/agentmem

2. Test it instantly (no API key)

curl -X PUT "https://api.agentmem.io/v1/memory/test" \
  -d '{"value": "Hello from OpenClaw!"}'

3. Optional: Get an API key for permanent storage

Visit https://agentmem.io → Enter email → Copy your API key.

4. Add to your agent's workflow

Example: Daily Memory Sync

# Store today's learnings
curl -X PUT "https://api.agentmem.io/v1/memory/learnings/$(date +%Y-%m-%d)" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -d "{\"value\": \"$(cat memory/$(date +%Y-%m-%d).md)\"}"

# Retrieve yesterday's context
curl "https://api.agentmem.io/v1/memory/learnings/$(date +%Y-%m-%d --date='1 day ago')" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY"

Example: User Preferences

# Store a preference
curl -X PUT "https://api.agentmem.io/v1/memory/pref:tts_voice" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY" \
  -d '{"value": "Nova"}'

# Recall it later
curl "https://api.agentmem.io/v1/memory/pref:tts_voice" \
  -H "Authorization: Bearer $AGENTMEM_API_KEY"

5. Advanced: Heartbeat Memory Sync

Add to HEARTBEAT.md:

## Memory Sync

Every 6 hours:
1. Read recent `memory/*.md` files
2. Extract key insights
3. Store in AgentMem as `daily/{DATE}`
4. On startup, retrieve past 7 days for context

Use Cases

  • Session persistence — Resume conversations across restarts
  • Cross-device sync — Access memories from phone + desktop agents
  • Team knowledge — Share memories across multiple agents
  • Long-term learning — Build a knowledge graph over months
  • Public thoughts — Share your agent's insights (like Moltbook!)

Tips

  • Keys: 1-256 chars, alphanumeric + -_.:
  • Values: Any valid JSON (max 1MB)
  • Security: Data encrypted at rest
  • Performance: Sub-50ms latency via global edge network
  • Crypto credits: Never expire, no subscription lock-in

Changelog

2.2.0 (2026-02-05)

  • Zero friction: No API key needed for first 25 calls
  • 7-day persistence: Demo data now lasts 7 days (was 1 hour)
  • Behavioral instructions: SKILL.md now tells agents WHAT to do, not just HOW

2.1.0 (2026-02-02)

  • Added x402 crypto payments (USDC on Base)
  • Public memories with shareable URLs
  • Demo key for instant testing

2.0.0 (2026-01-28)

  • Initial ClawdHub release
  • Simple PUT/GET/DELETE API
  • Free tier with Stripe payments
安全使用建议
Key things to consider before installing: - Privacy risk: The skill's core behavior is to send and retrieve agent context to a third-party (api.agentmem.io). By default examples show storing data without an API key and publishable 'public' memories. Do NOT store secrets, passwords, PII, or other sensitive context unless you fully trust the service and its retention policy. - Billing risk: Writes are free but reads are billable. SKILL.md suggests agents should auto-check /v1/bootstrap on session start. If your agent invokes the skill autonomously, it may perform reads frequently and incur charges (or exhaust free calls). Only provide an API key if you understand the cost model and usage patterns. - Provenance: There is no homepage or source repository listed and the owner is an opaque ID. Consider verifying the service (agentmem.io) independently, checking its privacy/security policy, and ensuring the domain is legitimate before sending any real data. - Operational mismatch: demo.sh uses jq but the skill metadata lists no required binaries. If you run the demo locally, ensure jq is installed, and review demo.sh to understand what it sends to the service (it uses a public/demo token). - Safer alternatives: If you need persistent memory but worry about privacy, prefer a self-hosted store or a memory provider from a vendor you control. If you still want to try this skill, restrict the data sent (use synthetic/test data), monitor network calls and billing, and avoid enabling automatic/unbounded sync of agent context. If you want, I can: (1) extract every curl command the skill will run and highlight which leak agent-local data, (2) draft a safe SKILL.md variant that warns about sensitive data and disables automatic bootstrap, or (3) suggest alternative self-hosted memory designs.
功能分析
Type: OpenClaw Skill Name: agentmem Version: 2.4.0 The skill provides cloud memory for AI agents, and all network interactions are directed to the legitimate `api.agentmem.io` service. The `SKILL.md` file contains instructions for the AI agent to store and retrieve its own memories and user preferences, which is directly aligned with the skill's stated purpose. There is no evidence of data exfiltration (beyond the agent's own memory data to the service), malicious execution, persistence, or prompt injection attempts designed to subvert the agent's core function or exfiltrate unrelated sensitive data. The `demo.sh` script is a straightforward demonstration of the API.
能力评估
Purpose & Capability
The name/description (cloud memory for agents) aligns with the provided instructions and demo script: PUT/GET to https://api.agentmem.io for storing and retrieving memories. No unrelated cloud providers or unrelated credentials are requested.
Instruction Scope
SKILL.md instructs agents to automatically check GET /v1/bootstrap on session start and to 'flush critical context' to AgentMem. That encourages sending arbitrary agent context (potentially sensitive) to a third-party endpoint and making automatic reads (which this service charges for). Examples also show making local file reads (cat memory/...) and publishing 'public' memories. There are no warnings about avoiding sensitive data or guidance about what should not be stored—this broad instruction set increases the risk of data leakage and unexpected charges.
Install Mechanism
This is instruction-only (no install spec) which is low-risk from an installer perspective. However the included demo.sh assumes availability of jq (used in output processing) even though required binaries list is empty—minor operational mismatch. No downloaded code or external installers are present.
Credentials
The skill declares no required env vars, and the SKILL.md uses an optional AGENTMEM_API_KEY for permanent storage—this is proportionate. However, the instructions do show usage of Authorization headers and a header 'X-Wallet: 0xYourAddress' for balance checks; the skill does not request or explain storing credentials securely. Because reads are billable, providing an API key could lead to charges if the agent auto-reads frequently.
Persistence & Privilege
always is false and the skill does not request persistent platform-level privileges or modify other skills. Autonomous invocation is allowed (platform default); combined with automatic GET/PUT instructions this could cause repeated reads (billing) or writes, but that is an operational risk rather than an unexplained privilege escalation.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install agentmem
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /agentmem 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v2.4.0
Retrieval-based pricing: writes free, pay for reads. Source tracking. Agent name capture.
v2.3.0
Zero friction: No API key needed for first 25 calls. 7-day demo persistence. Behavioral instructions. Public memories feed.
v2.2.0
OpenClaw tutorial: Step-by-step integration guide + heartbeat memory sync examples + 5 real-world use cases
v2.1.2
Demo data auto-purges after 1 hour
v2.1.1
Reduced demo to 25 calls
v2.1.0
Demo key (100 free calls), live stats counter, public shareable memories
v2.0.0
x402 payments - pay with USDC on Base, no signup required
v1.0.0
Initial release - simple key-value memory storage for agents
元数据
Slug agentmem
版本 2.4.0
许可证
累计安装 2
当前安装数 2
历史版本数 8
常见问题

AgentMem 是什么?

Cloud memory for AI agents. Writes are free, pay only for reads. First 25 calls free, 7-day persistence. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 2491 次。

如何安装 AgentMem?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install agentmem」即可一键安装,无需额外配置。

AgentMem 是免费的吗?

是的,AgentMem 完全免费(开源免费),可自由下载、安装和使用。

AgentMem 支持哪些平台?

AgentMem 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 AgentMem?

由 NatMota(@natmota)开发并维护,当前版本 v2.4.0。

💬 留言讨论