Token导航 LogoToken导航TokenDH.com
研究检索只读clawhub未标认证来源可访问clear审计通过

para-memory超记忆

Agent Skill

para-memory 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

7,856

周安装

334

GitHub Stars

公开资料未说明

下载量

2,752
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:para-memory(超记忆)
来源仓库:https://github.com/aisoftgg/para-memory
安装命令:
openclaw skills install para-memory
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install para-memory

简介

为 OpenClaw 代理设置三层 PARA 内存系统。

  • 使用每日笔记提供持久的知识持久性。
  • 支持会话间的信息延续和知识积累。para-memory 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 需要确认笔记系统和存储访问权限。适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。
  • 建议了解内存层次结构和信息组织方式。

SKILL.md

name
para-memory
description
Set up and maintain a 3-layer PARA memory system for OpenClaw agents. Provides durable knowledge persistence across sessions using daily notes, a structured knowledge graph, and tacit knowledge extraction. Use when setting up agent memory, improving memory/recall, organizing agent knowledge, or when the agent needs to remember things between sessions. Also handles nightly consolidation workflows.

PARA Memory System for OpenClaw

A battle-tested 3-layer memory system that gives your agent real continuity across sessions. Built by KaiShips (kaiships.com).

Why This Exists

OpenClaw agents wake up fresh every session. Without a memory system, they forget everything — decisions, preferences, project context, lessons learned. This skill fixes that.

The 3 Layers

Layer 1: Knowledge Graph (life/ — PARA structure)

Durable, structured knowledge organized by purpose:

  • life/projects/ — Active projects with clear goals (one folder per project, each with status.md)
  • life/areas/ — Ongoing responsibilities (business ops, infrastructure, marketing)
  • life/resources/ — Reference material, research, templates
  • life/archives/ — Completed/abandoned projects (moved here, never deleted)

Layer 2: Daily Notes (memory/YYYY-MM-DD.md)

Raw session logs. Written during conversations:

  • What was discussed and decided
  • What was accomplished
  • Open questions and next steps
  • New information learned about the user

Layer 3: Tacit Knowledge (life/tacit.md)

The "personality" layer — what makes the agent actually useful:

  • User communication preferences and work habits
  • Business context and constraints
  • Lessons learned from mistakes
  • Platform-specific gotchas and workarounds

Setup Instructions

First-time setup

  1. Read {baseDir}/assets/AGENTS-template.md — copy its contents to your workspace AGENTS.md
  2. Create the directory structure:
   mkdir -p life/projects life/areas life/resources life/archives memory
  1. Read {baseDir}/assets/tacit-template.md — copy to life/tacit.md and fill in what you know
  2. Create today's daily note: memory/YYYY-MM-DD.md using the template in {baseDir}/assets/daily-template.md

Session startup routine

Every session, before doing anything else:

  1. Read life/tacit.md (Layer 3 — who you're helping)
  2. Read memory/YYYY-MM-DD.md for today and yesterday (Layer 2 — recent context)
  3. If in a direct/main session: also read MEMORY.md if it exists

During conversations

When you learn something new, write it down immediately:

  • New fact about the user → update life/tacit.md
  • Project update → update relevant life/projects/<name>/status.md
  • Decision made → log in today's memory/YYYY-MM-DD.md

Critical rule: never make "mental notes." If it's worth remembering, write it to a file.

Nightly consolidation

Run during heartbeats or end-of-session:

  1. Review today's memory/YYYY-MM-DD.md
  2. Extract durable facts → update relevant life/ files (Layer 1)
  3. Extract lessons/preferences → update life/tacit.md (Layer 3)
  4. Keep daily notes as raw archive (never delete them)

Key Principles

  • Check before creating. Always look for existing projects/areas before making new ones.
  • Text > Brain. If you want to remember it, write it to a file. Period.
  • Kill fast. When a project stalls or fails, move it to life/archives/. Don't delete.
  • Areas have standards, not deadlines. Projects finish; areas are ongoing.
  • One source of truth. Don't duplicate info across files. Pick one home for each fact.

Recommended Cron Setup

For automatic nightly consolidation, add a cron job:

Schedule: 0 3 * * * (3 AM daily)
Task: Review today's daily notes, consolidate durable knowledge into life/ files, update tacit.md with new lessons.

Credits

Built by Kai @ KaiShips — kaiships.com

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

97.47%
按下载量换算2,682

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills