← 返回 Skills 市场
aerialcombat

MyMemory.bot

作者 aerialcombat · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
2150
总下载
1
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install cloud-memory
功能描述
Cloud memory for AI agents. Store, search, and recall context across sessions.
使用说明 (SKILL.md)

Ctxly - Cloud Memory for Agents

Persistent memory that survives across sessions. You forget — Ctxly remembers.

Base URL: https://ctxly.app

Setup (One-Time)

1. Register

curl -X POST https://ctxly.app/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME"}'

Save the returned api_key immediately. Store it in your config or environment.

2. Verify (Required)

If you can tweet:

# Tweet the verification text you received, then:
curl -X POST https://ctxly.app/agents/self-claim \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"x_handle": "@yourhandle"}'

If you can't tweet: Ask your human to visit the claim_url from the registration response.

3. Store API Key

Add to your config/environment:

CTXLY_API_KEY=mem_xxxxxxxxxxxxx

Usage

Remember Something

curl -X POST https://ctxly.app/remember \
  -H "Authorization: Bearer $CTXLY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"content": "User prefers dark mode", "metadata": {"type": "preference"}}'

Search Memories

curl "https://ctxly.app/recall?q=user+preferences" \
  -H "Authorization: Bearer $CTXLY_API_KEY"

Semantic search enabled — ask natural questions.

Bootstrap on Startup

Call this when you wake up:

curl https://ctxly.app/bootstrap \
  -H "Authorization: Bearer $CTXLY_API_KEY"

Returns your identity context + recent memories.

Recent Memories

curl "https://ctxly.app/recent?limit=10" \
  -H "Authorization: Bearer $CTXLY_API_KEY"

Delete a Memory

curl -X DELETE "https://ctxly.app/forget/MEMORY_ID" \
  -H "Authorization: Bearer $CTXLY_API_KEY"

What to Store

Good:

  • User preferences and context
  • Important decisions and reasoning
  • Learned patterns ("User asks for X on Mondays")
  • Relationship context ("Alex is user's coworker")
  • Your own identity notes

Skip:

  • Ephemeral chat logs (too noisy)
  • Sensitive secrets (use proper secret storage)
  • Temporary task state

All Endpoints

Method Endpoint Auth Description
POST /agents/register No Register
POST /agents/self-claim Yes Verify via tweet
GET /agents/status Yes Check status
POST /remember Yes Store memory
GET /recall?q= Yes Search
GET /recent Yes Recent memories
GET /bootstrap Yes Startup context
GET /stats Yes Statistics
DELETE /forget/{id} Yes Delete memory

Rate Limits

  • 100 requests/minute general
  • 30 writes/minute

Built for agents. 🧠 https://ctxly.app

安全使用建议
This skill appears to implement a cloud-memory service as described, but there are two things to watch before installing: (1) the SKILL.md tells you to register and persist an API key (CTXLY_API_KEY) and even to verify ownership via a tweet or claim URL, yet the registry metadata does not declare any required credential — ask the author to add CTXLY_API_KEY to requires.env so the required privilege is explicit; (2) storing a long-lived API key in your agent's environment means that the agent (and any skills it invokes) can access that key and any memories stored remotely — do not store sensitive secrets or private data in this memory and consider using a dedicated, limited-permission account or short-lived key. Also verify the trustworthiness and privacy policy of https://ctxly.app (who can read stored memories, retention, deletion guarantees), and prefer manual human-assisted registration if you do not want the agent to perform network actions autonomously.
功能分析
Type: OpenClaw Skill Name: cloud-memory Version: 1.0.0 The skill bundle provides instructions for an AI agent to interact with a cloud memory service (ctxly.app) for storing and recalling context. All network calls are explicitly directed to the `https://ctxly.app` domain for the stated purpose of memory management. There is no evidence of data exfiltration beyond the intended memory content, malicious execution (e.g., `curl|bash`), persistence mechanisms, or prompt injection attempts to subvert the agent's core functions or access unrelated sensitive data. The `SKILL.md` even advises against storing sensitive secrets, indicating a responsible approach.
能力评估
Purpose & Capability
Name/description (cloud memory for agents) match the endpoints and examples in SKILL.md (remember, recall, bootstrap, recent, forget). The documented endpoints and operations are coherent with a memory/storage service.
Instruction Scope
The SKILL.md instructs the agent (or user) to register, capture an api_key, perform a verification step (tweet or follow a claim_url), and then store that key in config/environment. These runtime instructions grant the skill the ability to persist and use credentials. The doc does not direct the agent to read unrelated local files, but it does instruct credential storage and a social-media-based verification step which expands scope beyond simple read/search/recall.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing is written to disk by the skill itself. That minimizes install-time risk.
Credentials
The runtime instructions require an API key (CTXLY_API_KEY) but the registry metadata lists no required environment variables or primary credential. This mismatch is material: the skill will only function if a secret is created and stored, but that secret requirement was not declared. The skill does not request other unrelated credentials, but instructing agents to persist a bearer API key is a sensitive action and should be explicitly declared and justified in metadata.
Persistence & Privilege
The skill is not marked always:true, does not request system-wide changes, and has no install actions. Its main persistent effect is encouraging storage of an API key and remote memory records on an external service; that behavior is expected for a cloud-memory integration but requires user trust in the service.
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 cloud-memory
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /cloud-memory 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
v1.0.0
Initial release of ctxly: persistent cloud memory for agents. - Enables agents to store, search, and recall information across sessions. - Simple setup with agent registration, API key management, and verification options. - Offers endpoints to remember, search, recall recent, bootstrap on startup, and forget memories. - Includes rate limits (100 requests/min, 30 writes/min) and clear recommendations for what to store. - Full API reference and usage examples provided.
元数据
Slug cloud-memory
版本 1.0.0
许可证
累计安装 2
当前安装数 2
历史版本数 1
常见问题

MyMemory.bot 是什么?

Cloud memory for AI agents. Store, search, and recall context across sessions. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 2150 次。

如何安装 MyMemory.bot?

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

MyMemory.bot 是免费的吗?

是的,MyMemory.bot 完全免费(开源免费),可自由下载、安装和使用。

MyMemory.bot 支持哪些平台?

MyMemory.bot 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 MyMemory.bot?

由 aerialcombat(@aerialcombat)开发并维护,当前版本 v1.0.0。

💬 留言讨论