← 返回 Skills 市场
MemData
作者
thelabvenice
· GitHub ↗
· v1.8.0
1612
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install memdata
功能描述
Persistent memory for autonomous agents. Wallet = identity. Pay per query. Optional encrypted storage.
安全使用建议
This skill appears coherent, but before installing: 1) Verify the service (memdata.ai) and read its docs and privacy policy. 2) Confirm how your agent will sign x402 payments and UCAN delegations — do not allow silent programmatic access to wallet private keys; require interactive signing. 3) Test with a tiny payment to confirm behavior and pricing. 4) If you plan to use encrypted mode, understand the UCAN delegation flow and where the delegationCar is stored. 5) Consider whether you want the agent to be allowed to call this skill autonomously; if not, disable model invocation or require user confirmation for actions that initiate payments.
功能分析
Type: OpenClaw Skill
Name: memdata
Version: 1.8.0
The skill bundle describes a legitimate service for persistent memory for autonomous agents, interacting with `https://memdata.ai/api/x402` via a defined set of API endpoints and an x402 payment protocol. The `SKILL.md` file serves as documentation for using this service and contains no evidence of prompt injection attempts against the AI agent, data exfiltration, malicious execution, persistence mechanisms, or obfuscation. All described actions are aligned with the stated purpose of providing memory functionality.
能力评估
Purpose & Capability
Name/description (persistent memory, wallet identity, pay-per-query) match the SKILL.md: endpoints, pricing, and encrypted-storage flow are all aligned with that purpose. No unrelated env vars, binaries, or install steps are requested.
Instruction Scope
SKILL.md is an instruction-only API spec that tells the agent to call memdata.ai endpoints and to perform a wallet-based x402 payment flow. It does not instruct reading local files or environment variables. It omits low-level details about how to produce the signed payment header or UCAN delegation (it assumes the agent/user will sign via a wallet), which is expected but important to understand.
Install Mechanism
No install spec and no code files — instruction-only skill. This minimizes on-disk risk; nothing is downloaded or installed by the skill itself.
Credentials
The skill declares no required environment variables or credentials, which is consistent with a wallet-based, on-demand signing model. However, practical use requires a wallet/signing capability (private key or wallet integration). The SKILL.md does not request keys directly, so ensure your agent/wallet does not expose private keys to the skill without explicit consent.
Persistence & Privilege
No special persistence flags (always not set). Model invocation is not disabled (default allows autonomous invocation). That is acceptable for a network-based memory API, but because endpoints involve payments, confirm the agent cannot programmatically sign transactions or use stored private keys without explicit interactive approval.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install memdata - 安装完成后,直接呼叫该 Skill 的名称或使用
/memdata触发 - 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
v1.8.0
MemData 1.8.0 introduces persistent memory with wallet-based identity and optional encrypted storage for autonomous agents. Key updates:
- No registration or API keys required; payments (USDC on Base) manage authentication and identity.
- Endpoints for ingesting, querying, managing identity, encrypted storage setup, and artifact management.
- Transparent per-endpoint pricing; queries start at $0.001.
- Optional end-to-end encryption via Lit Protocol and IPFS/Filecoin backend.
- Query results include memory quality grounding for better context on data depth and reliability.
元数据
常见问题
MemData 是什么?
Persistent memory for autonomous agents. Wallet = identity. Pay per query. Optional encrypted storage. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1612 次。
如何安装 MemData?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install memdata」即可一键安装,无需额外配置。
MemData 是免费的吗?
是的,MemData 完全免费(开源免费),可自由下载、安装和使用。
MemData 支持哪些平台?
MemData 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 MemData?
由 thelabvenice(@thelabvenice)开发并维护,当前版本 v1.8.0。
推荐 Skills