mmagMixed Memory-Augmented Generation (MMAG) for AI agents. Five cognitive memory layers — conversational, long-term user, episodic, sensory, and short-term work...
Install via ClawdBot CLI:
clawdbot install j0ker98/mmagGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Potentially destructive shell commands in tool definitions
eval (Uses known external API (expected, informational)
arxiv.orgAudited Apr 17, 2026 · audit v1.0
Generated Mar 20, 2026
A customer service AI that remembers past interactions, user preferences, and issues across sessions to provide tailored, consistent support without repeating questions. It uses long-term memory for user profiles and episodic memory for tracking recurring problems or follow-ups.
An AI assistant that tracks patient-reported symptoms, medication schedules, and lifestyle habits over time using episodic and sensory memory layers. It personalizes health reminders and adapts advice based on long-term user data, improving adherence and outcomes.
An agent that integrates with IoT devices to remember user routines, environmental preferences (like temperature or lighting), and scheduled events. It uses sensory memory for real-time context and long-term memory for habit optimization, enhancing energy efficiency and comfort.
A tutoring AI that recalls a student's learning history, strengths, weaknesses, and progress across sessions. It leverages conversational memory for session coherence and long-term memory to personalize lesson plans, making education more effective and engaging.
An AI tool that tracks project timelines, team interactions, and task dependencies using episodic and working memory layers. It helps prioritize tasks based on past events and sensory context (like deadlines), improving productivity and coordination in remote teams.
Offer MMAG as a cloud API where developers pay monthly or annually for memory storage, retrieval, and processing tiers. Revenue scales with usage, such as number of memory layers accessed or data volume, targeting businesses needing persistent AI agents.
Sell licenses to large organizations for self-hosted MMAG installations, ensuring data privacy and customization. This includes support, training, and updates, appealing to industries like healthcare or finance with strict compliance requirements.
Provide a free version with basic memory layers and limited storage, while charging for advanced features like encryption, high-volume snapshots, or priority support. This attracts individual developers and small teams, converting them to paid plans as needs grow.
💬 Integration Tip
Integrate MMAG by running context.sh at session start to inject memory into the system prompt, and use store.sh scripts during interactions to update layers based on triggers like user preferences or events.
Scored Apr 21, 2026
智能客服机器人 - 24小时在线客服，自动回复、知识库管理、多渠道支持
切换 agent 的人设（soul.md）。支持三套预设人设与原有人设之间自由切换。 触发词：切换人设、persona-switch、赛博朋友、温柔伴侣、创始人龙虾。
Cross-functional organizational health check combining signals from all C-suite roles. Scores 8 dimensions on a traffic-light scale with drill-down recommend...
Build React chat interfaces with Vercel AI SDK v6. Covers useChat/useCompletion/useObject hooks, message parts structure, tool approval workflows, and 18 UI error solutions. Prevents documented issues with React Strict Mode, concurrent requests, stale closures, and tool approval edge cases. Use when: implementing AI chat UIs, migrating v5→v6, troubleshooting "useChat failed to parse stream", "stale body values", "React maximum update depth", "Cannot read properties of undefined (reading 'state')", or tool approval workflow errors.
Print a friendly greeting. Use when the user asks to be greeted, needs a simple hello demo, or wants a no-network text output test.
Builds consensus-driven AI personas from multiple perspectives