跳到正文
TakoAPI
返回技能列表

mindgardener

由 widingmarcus-cyber 提供

Coding Agents & IDEs
489下载量由 clawskills.sh 提供
2星标
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

Local-first long-term memory for autonomous agents.

描述

该技能的功能

Local-first long-term memory for OpenClaw agents. Extracts entities from conversations into a wiki knowledge graph, scores events by surprise value, detects conflicts between new and old information, and assembles token-budget context. Everything is stored as markdown files with no database or external API required.

Unlike OpenClaw's built-in memory_search, which only queries existing memory, MindGardener creates and structures memory from conversations, adds surprise scoring, conflict detection, and graph traversal.

何时使用

  • Persisting project facts across multiple agent sessions
  • Flagging when new information contradicts stored knowledge
  • Building a navigable knowledge graph from conversation history
  • Injecting relevant context automatically at session start
  • Syncing per-agent memories to a shared store in multi-agent workflows

示例工作流

输入

User asks: remember everything from today's planning session

智能体
  1. 1

    Run garden extract to pull entities and facts from the conversation

  2. 2

    Run garden surprise to score unexpected or high-value facts

  3. 3

    Run garden consolidate to merge new facts into the knowledge graph and detect conflicts

  4. 4

    Run garden inject --output RECALL-CONTEXT.md at the start of the next session to assemble relevant context

输出

RECALL-CONTEXT.md populated with relevant facts within the token budget, ready to inject into the agent's context window

安装

clawhub install widingmarcus-cyber/mindgardener

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接