mindgardenerLocal-first long-term memory for autonomous agents. Built for OpenClaw. Creates wiki knowledge graph from conversations, scores events by surprise, detects c...
Install via ClawdBot CLI:
clawdbot install widingmarcus-cyber/mindgardenerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Potentially destructive shell commands in tool definitions
eval (Calls external URL not in known-safe list
https://github.com/widingmarcus-cyber/mindgardenerUses known external API (expected, informational)
api.anthropic.comAI Analysis
The skill is a local-first memory tool with no evidence of unauthorized data exfiltration, credential harvesting, or hidden malicious instructions. The primary risk is the potential for unsafe shell command execution (eval) and reliance on an external GitHub repository for installation, which is typical for open-source tools.
Audited Apr 16, 2026 · audit v1.0
Generated Mar 20, 2026
Enhances AI support agents by automatically creating a knowledge graph from past conversations, enabling accurate and context-aware responses. Detects conflicting information to maintain consistency and scores surprising events to prioritize important customer issues.
Facilitates multi-agent AI systems in R&D teams by syncing memories across agents, tracking provenance of facts, and detecting conflicts in experimental data. Assembles token-budget context for efficient information retrieval during collaborative projects.
Assists legal professionals by extracting entities and relationships from case discussions into a wiki knowledge graph. Uses conflict detection to flag inconsistencies in legal precedents and provides associative recall for quick reference during case preparation.
Integrates with AI agents in healthcare to create long-term memory from patient interactions, scoring unexpected symptoms for importance. Ensures data consistency with conflict detection and generates context for personalized care plans without external APIs.
Supports AI tutors by building a knowledge graph from student conversations, highlighting surprising learning gaps. Uses temporal decay to focus on recent topics and provides auto-injected context for adaptive lesson planning in offline environments.
Offer MindGardener as a free, open-source tool for developers, with paid tiers for enterprise support, custom integrations, and advanced features like enhanced conflict resolution. Revenue comes from subscription fees and consulting services.
Provide a cloud-based version with additional analytics, collaboration tools, and automated backups, targeting teams using OpenClaw. Charge based on usage metrics such as memory size or number of agents, with a free tier for small projects.
Sell licenses to corporations for integrating MindGardener into their proprietary AI systems, with features like multi-agent sync at scale and custom deployment options. Revenue is generated through one-time licenses and annual maintenance fees.
💬 Integration Tip
Install via pip and run garden init to set up local files; schedule garden commands in cron for automated memory updates and use garden inject to generate context at session start.
Scored Apr 21, 2026
Structured reasoning through sequential thinking — break complex problems into steps, solve each independently, verify consistency, synthesize conclusions wi...
LLM-driven epistemic reasoning engine. Evaluates claims against evidence, outputs calibrated confidence and structured belief state (VERIFIED/CONTESTED/UNCER...
Loads and manages company context for all C-suite advisor skills. Reads ~/.claude/company-context.md, detects stale context (>90 days), enriches context duri...
Turn scattered local sources into a source-constrained evidence notebook for incident, release, and maintainer decisions.
三级记忆管理系统 (Three-Tier Memory Management)。用于管理 AI 代理的短期、中期、长期记忆。包括:(1) 滑动窗口式短期记忆,(2) 自动摘要生成中期记忆,(3) 向量检索长期记忆 (RAG)。当需要管理对话历史、优化上下文、构建个人知识库、或实现记忆持久化时使用此 Skill。
Store secrets, long-term memory, daily logs, and anything custom in your Convex backend instead of local files