AgentMem

Cloud memory for AI agents. Writes are free, pay only for reads. First 25 calls free, 7-day persistence.

Security Scan

The skill appears to implement a cloud memory service consistent with its name, but its runtime instructions encourage sending arbitrary agent context to an external service (including public writes), lacks provenance (no homepage/source), and has a few inconsistencies (demo script uses jq) that raise privacy, billing, and operational concerns you should understand before installing.

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.

What to consider before installing

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.

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