Merge bac74ca9d6 into 92568c1eda

README.md

@@ -127,6 +127,7 @@ Skills are not MCP servers and not tools. MCP defines how an agent connects to e
 - [Full-Page Screenshot](https://github.com/LewisLiu007/full-page-screenshot) - Captures full-page screenshots of web pages via Chrome DevTools Protocol with zero dependencies. *By [@LewisLiu007](https://github.com/LewisLiu007)*
 - [great_cto](https://github.com/avelikiy/great_cto) - Claude Code plugin: 7 specialised subagents (tech-lead, senior-dev, qa-engineer, security-officer, devops, l3-support, project-auditor) orchestrating a full SDLC pipeline — architecture, TDD, 12-angle code review, QA, security audit, deploy. 11 project archetypes auto-detected, 13 compliance frameworks (GDPR/PCI-DSS/HIPAA/SOC2/ISO 27001), self-improving knowledge layer that learns from every incident. *By [@avelikiy](https://github.com/avelikiy)*
 - [iOS Simulator](https://github.com/conorluddy/ios-simulator-skill) - Enables Claude to interact with iOS Simulator for testing and debugging iOS applications. *By [@conorluddy](https://github.com/conorluddy)*
+- [Jarvis Orb](https://github.com/TheStack-ai/jarvis-orb) - Persistent 4-tier AI memory with temporal scoring, contradiction detection, entity tracking, and real-time desktop visualization orb via MCP. *By [@TheStack-ai](https://github.com/TheStack-ai)*
 - [jules](https://github.com/sanjay3290/ai-skills/tree/main/skills/jules) - Delegate coding tasks to Google Jules AI agent for async bug fixes, documentation, tests, and feature implementation on GitHub repos. *By [@sanjay3290](https://github.com/sanjay3290)*
 - [LangSmith Fetch](./langsmith-fetch/) - Debug LangChain and LangGraph agents by automatically fetching and analyzing execution traces from LangSmith Studio. First AI observability skill for Claude Code. *By [@OthmanAdi](https://github.com/OthmanAdi)*
 - [lean-ctx](https://github.com/yvgude/lean-ctx) - MCP server and context runtime for AI coding agents: session caching, AST-aware compression, and 90+ shell patterns to reduce token usage. Supports Claude Code, Cursor, Copilot, and other integrations. Install the Claude Code skill with `lean-ctx init --agent claude-code`; docs at [leanctx.com](https://leanctx.com). *By [@yvgude](https://github.com/yvgude)*

jarvis-orb/SKILL.md

@@ -0,0 +1,66 @@
+---
+name: jarvis-orb
+description: Persistent 4-tier AI memory system with temporal scoring, contradiction detection, entity tracking, and real-time desktop visualization via MCP.
+---
+
+# Jarvis Orb
+
+A persistent memory layer for AI coding assistants that remembers context across sessions. Provides 4-tier memory (episodic, semantic, project, procedural) with temporal relevance scoring, automatic contradiction detection, and entity relationship tracking. Includes a real-time desktop visualization orb that shows the AI's thinking process.
+
+## When to Use This Skill
+
+- When you need persistent memory across Claude Code sessions
+- When working on long-running projects that require context retention
+- When you want automatic entity tracking and relationship mapping
+- When you need contradiction detection in stored knowledge
+- When you want real-time visualization of AI memory operations
+
+## What This Skill Does
+
+1. **4-Tier Memory**: Organizes knowledge into episodic (events), semantic (facts), project (context), and procedural (how-to) tiers
+2. **Temporal Scoring**: Automatically ranks memory relevance based on recency, frequency, and importance
+3. **Contradiction Detection**: Identifies and resolves conflicting information across memory tiers
+4. **Entity Tracking**: Maps relationships between people, projects, decisions, and concepts
+5. **Desktop Visualization**: Real-time orb animation showing memory read/write operations
+
+## How to Use
+
+### Setup
+
+**macOS (Apple Silicon) / Linux:**
+```bash
+curl -fsSL https://raw.githubusercontent.com/TheStack-ai/jarvis-orb/main/install.sh | bash
+```
+
+**Windows (PowerShell):**
+```powershell
+irm https://raw.githubusercontent.com/TheStack-ai/jarvis-orb/main/install.ps1 | iex
+```
+
+The installer configures the MCP server and launches the desktop orb automatically.
+
+### Basic Usage
+
+Once configured as an MCP server, Jarvis Orb automatically:
+- Stores important context from conversations
+- Retrieves relevant memories when needed
+- Tracks entities and their relationships
+- Displays real-time activity via the desktop orb
+
+## Compatibility
+
+- Claude Code
+- Cursor
+- Any MCP-compatible AI tool
+
+## Tips
+
+- Start with project-level memory for focused context retention
+- Use the visualization orb to monitor memory operations in real-time
+- Review contradiction alerts to keep knowledge base consistent
+
+## Common Use Cases
+
+- Multi-session project development with full context retention
+- Team knowledge management across AI-assisted workflows
+- Long-term decision tracking and rationale preservation