agent-os-kernel 1.2.0__tar.gz → 1.3.0__tar.gz

{agent_os_kernel-1.2.0 → agent_os_kernel-1.3.0}/.gitignore

@@ -152,5 +152,8 @@ benchmark_results/
 # Experiment outputs
 experiments/outputs/
 experiments/logs/
-
+
 node_modules/
+
+# MCP Registry tokens
+.mcpregistry_*

{agent_os_kernel-1.2.0 → agent_os_kernel-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-os-kernel
-Version: 1.2.0
+Version: 1.3.0
 Summary: A kernel architecture for governing autonomous AI agents with Nexus Trust Exchange
 Project-URL: Homepage, https://github.com/imran-siddique/agent-os
 Project-URL: Documentation, https://github.com/imran-siddique/agent-os/tree/master/docs
@@ -70,12 +70,15 @@ Description-Content-Type: text/markdown
 
 **A kernel architecture for governing autonomous AI agents**
 
+[![GitHub Stars](https://img.shields.io/github/stars/imran-siddique/agent-os?style=social)](https://github.com/imran-siddique/agent-os/stargazers)
 [![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 [![Python](https://img.shields.io/badge/python-3.10+-blue.svg)](https://python.org)
 [![CI](https://github.com/imran-siddique/agent-os/actions/workflows/ci.yml/badge.svg)](https://github.com/imran-siddique/agent-os/actions/workflows/ci.yml)
 [![VS Code Extension](https://img.shields.io/badge/VS%20Code-Extension-007ACC?logo=visual-studio-code)](https://marketplace.visualstudio.com/items?itemName=agent-os.agent-os-vscode)
 [![Documentation](https://img.shields.io/badge/docs-imran--siddique.github.io-blue)](https://imran-siddique.github.io/agent-os-docs/)
 
+> ⭐ **If this project helps you, please star it!** It helps others discover Agent OS.
+
 [Quick Start](#quick-example) • [Documentation](https://imran-siddique.github.io/agent-os-docs/) • [VS Code Extension](https://marketplace.visualstudio.com/items?itemName=agent-os.agent-os-vscode) • [Examples](examples/)
 
 <br/>
@@ -152,6 +155,46 @@ This is the same principle operating systems use: applications request resources
 
 ## Architecture
 
+```mermaid
+graph TB
+    subgraph "Layer 4: Execution"
+        SCAK[Self-Correcting Agent Kernel]
+        MUTE[Mute Agent]
+        ATR[Agent Tool Registry]
+    end
+    
+    subgraph "Layer 3: Control Plane"
+        KERNEL[🎯 THE KERNEL<br/>Policy Engine + Signals]
+        OBS[Observability<br/>Prometheus + OTEL]
+    end
+    
+    subgraph "Layer 2: Communication"
+        AMB[Agent Message Bus]
+        IATP[Inter-Agent Trust Protocol]
+        CMVK[Cross-Model Verification]
+        EMK[Episodic Memory Kernel]
+    end
+    
+    subgraph "Layer 1: Primitives"
+        PRIM[Base Types + Failures]
+        CAAS[Context-as-a-Service]
+    end
+    
+    SCAK --> KERNEL
+    MUTE --> KERNEL
+    ATR --> KERNEL
+    KERNEL --> AMB
+    KERNEL --> IATP
+    KERNEL --> OBS
+    AMB --> PRIM
+    IATP --> CMVK
+    CMVK --> EMK
+    EMK --> PRIM
+    CAAS --> PRIM
+```
+
+### Directory Structure
+
 ```
 agent-os/
 ├── src/agent_os/             # Core Python package
@@ -172,9 +215,9 @@ agent-os/
 │   ├── mute-agent/           # Layer 4: Face/Hands architecture
 │   └── mcp-kernel-server/    # Integration: MCP protocol support
 ├── extensions/               # IDE & AI Assistant Extensions
+│   ├── mcp-server/           # ⭐ MCP Server (Copilot, Claude, Cursor)
 │   ├── vscode/               # VS Code extension
-│   ├── mcp-server/           # Claude Desktop MCP Server
-│   ├── copilot/              # GitHub Copilot integration
+│   ├── copilot/              # GitHub Copilot (DEPRECATED - use mcp-server)
 │   ├── jetbrains/            # IntelliJ/PyCharm plugin
 │   ├── cursor/               # Cursor IDE extension
 │   ├── chrome/               # Chrome extension
@@ -215,11 +258,11 @@ agent-os/
 
 | Extension | Description |
 |-----------|-------------|
+| [`mcp-server`](extensions/mcp-server/) | ⭐ **MCP Server** - Works with Claude, Copilot, Cursor (`npx agentos-mcp-server`) |
 | [`vscode`](extensions/vscode/) | VS Code extension with real-time policy checks |
 | [`jetbrains`](extensions/jetbrains/) | IntelliJ, PyCharm, WebStorm plugin |
 | [`cursor`](extensions/cursor/) | Cursor IDE extension (Composer integration) |
-| [`copilot`](extensions/copilot/) | GitHub Copilot safety layer |
-| [`mcp-server`](extensions/mcp-server/) | **NEW** MCP Server for Claude Desktop |
+| [`copilot`](extensions/copilot/) | ~~GitHub Copilot safety layer~~ (DEPRECATED - use mcp-server) |
 | [`github-cli`](extensions/github-cli/) | `gh agent-os` CLI extension |
 | [`chrome`](extensions/chrome/) | Chrome extension for web agents |
 
@@ -470,31 +513,35 @@ agentos init
 
 ---
 
-## MCP Integration (Claude Desktop)
+## MCP Integration (Claude Desktop, GitHub Copilot, Cursor)
 
-Agent OS provides an MCP server for Claude Desktop integration:
+Agent OS provides an MCP server that works with any MCP-compatible AI assistant:
 
 ```bash
-# Install
-pip install agent-os[mcp]
+# Quick install via npx
+npx agentos-mcp-server
+```
+
+**npm:** [`agentos-mcp-server`](https://www.npmjs.com/package/agentos-mcp-server)  
+**MCP Registry:** `io.github.imran-siddique/agentos`
 
-# Run MCP server
-mcp-kernel-server --stdio
+Add to your config file:
 
-# Or add to claude_desktop_config.json:
+**Claude Desktop** (`%APPDATA%\Claude\claude_desktop_config.json` on Windows):
+```json
 {
   "mcpServers": {
-    "agent-os": {
-      "command": "mcp-kernel-server",
-      "args": ["--stdio"]
+    "agentos": {
+      "command": "npx",
+      "args": ["-y", "agentos-mcp-server"]
     }
   }
 }
 ```
 
-Exposes tools: `cmvk_verify`, `kernel_execute`, `iatp_sign`, `iatp_verify`
+**Features:** 10 tools for agent creation, policy enforcement, compliance checking (SOC 2, GDPR, HIPAA), human-in-the-loop approvals, and audit logging.
 
-See [MCP server documentation](modules/mcp-kernel-server/README.md).
+See [MCP server documentation](extensions/mcp-server/README.md) for full details.
 
 ---
 
@@ -602,24 +649,6 @@ export ANTHROPIC_API_KEY=sk-ant-...
 
 ---
 
-## FAQ
-
-**Q: How is this different from LangChain/CrewAI?**
-A: LangChain and CrewAI are *frameworks* for building agents. Agent OS is *infrastructure* for governing them. Use them together—wrap your LangChain agent with Agent OS for safety guarantees.
-
-**Q: What does "deterministic enforcement" mean?**
-A: When a policy matches an action, that action is blocked—not by asking the LLM nicely. The middleware intercepts and stops it. However, this only works for patterns the policy engine knows about. Novel attacks that don't match defined rules will pass through.
-
-**Q: Do I need to rewrite my agents?**
-A: No. Agent OS provides integration wrappers for LangChain, CrewAI, AutoGen, OpenAI Assistants, and Semantic Kernel. Wrap your existing code and add governance.
-
-**Q: Does it work with local models (Ollama, llama.cpp)?**
-A: Yes. Agent OS is model-agnostic—it governs what agents *do*, not what LLM they use.
-
-**Q: How do I contribute?**
-A: See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines. Good first issues are labeled in GitHub.
-
----
 
 ## Contributing
 

{agent_os_kernel-1.2.0 → agent_os_kernel-1.3.0}/README.md

@@ -4,12 +4,15 @@
 
 **A kernel architecture for governing autonomous AI agents**
 
+[![GitHub Stars](https://img.shields.io/github/stars/imran-siddique/agent-os?style=social)](https://github.com/imran-siddique/agent-os/stargazers)
 [![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 [![Python](https://img.shields.io/badge/python-3.10+-blue.svg)](https://python.org)
 [![CI](https://github.com/imran-siddique/agent-os/actions/workflows/ci.yml/badge.svg)](https://github.com/imran-siddique/agent-os/actions/workflows/ci.yml)
 [![VS Code Extension](https://img.shields.io/badge/VS%20Code-Extension-007ACC?logo=visual-studio-code)](https://marketplace.visualstudio.com/items?itemName=agent-os.agent-os-vscode)
 [![Documentation](https://img.shields.io/badge/docs-imran--siddique.github.io-blue)](https://imran-siddique.github.io/agent-os-docs/)
 
+> ⭐ **If this project helps you, please star it!** It helps others discover Agent OS.
+
 [Quick Start](#quick-example) • [Documentation](https://imran-siddique.github.io/agent-os-docs/) • [VS Code Extension](https://marketplace.visualstudio.com/items?itemName=agent-os.agent-os-vscode) • [Examples](examples/)
 
 <br/>
@@ -86,6 +89,46 @@ This is the same principle operating systems use: applications request resources
 
 ## Architecture
 
+```mermaid
+graph TB
+    subgraph "Layer 4: Execution"
+        SCAK[Self-Correcting Agent Kernel]
+        MUTE[Mute Agent]
+        ATR[Agent Tool Registry]
+    end
+    
+    subgraph "Layer 3: Control Plane"
+        KERNEL[🎯 THE KERNEL<br/>Policy Engine + Signals]
+        OBS[Observability<br/>Prometheus + OTEL]
+    end
+    
+    subgraph "Layer 2: Communication"
+        AMB[Agent Message Bus]
+        IATP[Inter-Agent Trust Protocol]
+        CMVK[Cross-Model Verification]
+        EMK[Episodic Memory Kernel]
+    end
+    
+    subgraph "Layer 1: Primitives"
+        PRIM[Base Types + Failures]
+        CAAS[Context-as-a-Service]
+    end
+    
+    SCAK --> KERNEL
+    MUTE --> KERNEL
+    ATR --> KERNEL
+    KERNEL --> AMB
+    KERNEL --> IATP
+    KERNEL --> OBS
+    AMB --> PRIM
+    IATP --> CMVK
+    CMVK --> EMK
+    EMK --> PRIM
+    CAAS --> PRIM
+```
+
+### Directory Structure
+
 ```
 agent-os/
 ├── src/agent_os/             # Core Python package
@@ -106,9 +149,9 @@ agent-os/
 │   ├── mute-agent/           # Layer 4: Face/Hands architecture
 │   └── mcp-kernel-server/    # Integration: MCP protocol support
 ├── extensions/               # IDE & AI Assistant Extensions
+│   ├── mcp-server/           # ⭐ MCP Server (Copilot, Claude, Cursor)
 │   ├── vscode/               # VS Code extension
-│   ├── mcp-server/           # Claude Desktop MCP Server
-│   ├── copilot/              # GitHub Copilot integration
+│   ├── copilot/              # GitHub Copilot (DEPRECATED - use mcp-server)
 │   ├── jetbrains/            # IntelliJ/PyCharm plugin
 │   ├── cursor/               # Cursor IDE extension
 │   ├── chrome/               # Chrome extension
@@ -149,11 +192,11 @@ agent-os/
 
 | Extension | Description |
 |-----------|-------------|
+| [`mcp-server`](extensions/mcp-server/) | ⭐ **MCP Server** - Works with Claude, Copilot, Cursor (`npx agentos-mcp-server`) |
 | [`vscode`](extensions/vscode/) | VS Code extension with real-time policy checks |
 | [`jetbrains`](extensions/jetbrains/) | IntelliJ, PyCharm, WebStorm plugin |
 | [`cursor`](extensions/cursor/) | Cursor IDE extension (Composer integration) |
-| [`copilot`](extensions/copilot/) | GitHub Copilot safety layer |
-| [`mcp-server`](extensions/mcp-server/) | **NEW** MCP Server for Claude Desktop |
+| [`copilot`](extensions/copilot/) | ~~GitHub Copilot safety layer~~ (DEPRECATED - use mcp-server) |
 | [`github-cli`](extensions/github-cli/) | `gh agent-os` CLI extension |
 | [`chrome`](extensions/chrome/) | Chrome extension for web agents |
 
@@ -404,31 +447,35 @@ agentos init
 
 ---
 
-## MCP Integration (Claude Desktop)
+## MCP Integration (Claude Desktop, GitHub Copilot, Cursor)
 
-Agent OS provides an MCP server for Claude Desktop integration:
+Agent OS provides an MCP server that works with any MCP-compatible AI assistant:
 
 ```bash
-# Install
-pip install agent-os[mcp]
+# Quick install via npx
+npx agentos-mcp-server
+```
+
+**npm:** [`agentos-mcp-server`](https://www.npmjs.com/package/agentos-mcp-server)  
+**MCP Registry:** `io.github.imran-siddique/agentos`
 
-# Run MCP server
-mcp-kernel-server --stdio
+Add to your config file:
 
-# Or add to claude_desktop_config.json:
+**Claude Desktop** (`%APPDATA%\Claude\claude_desktop_config.json` on Windows):
+```json
 {
   "mcpServers": {
-    "agent-os": {
-      "command": "mcp-kernel-server",
-      "args": ["--stdio"]
+    "agentos": {
+      "command": "npx",
+      "args": ["-y", "agentos-mcp-server"]
     }
   }
 }
 ```
 
-Exposes tools: `cmvk_verify`, `kernel_execute`, `iatp_sign`, `iatp_verify`
+**Features:** 10 tools for agent creation, policy enforcement, compliance checking (SOC 2, GDPR, HIPAA), human-in-the-loop approvals, and audit logging.
 
-See [MCP server documentation](modules/mcp-kernel-server/README.md).
+See [MCP server documentation](extensions/mcp-server/README.md) for full details.
 
 ---
 
@@ -536,24 +583,6 @@ export ANTHROPIC_API_KEY=sk-ant-...
 
 ---
 
-## FAQ
-
-**Q: How is this different from LangChain/CrewAI?**
-A: LangChain and CrewAI are *frameworks* for building agents. Agent OS is *infrastructure* for governing them. Use them together—wrap your LangChain agent with Agent OS for safety guarantees.
-
-**Q: What does "deterministic enforcement" mean?**
-A: When a policy matches an action, that action is blocked—not by asking the LLM nicely. The middleware intercepts and stops it. However, this only works for patterns the policy engine knows about. Novel attacks that don't match defined rules will pass through.
-
-**Q: Do I need to rewrite my agents?**
-A: No. Agent OS provides integration wrappers for LangChain, CrewAI, AutoGen, OpenAI Assistants, and Semantic Kernel. Wrap your existing code and add governance.
-
-**Q: Does it work with local models (Ollama, llama.cpp)?**
-A: Yes. Agent OS is model-agnostic—it governs what agents *do*, not what LLM they use.
-
-**Q: How do I contribute?**
-A: See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines. Good first issues are labeled in GitHub.
-
----
 
 ## Contributing
 

agent_os_kernel-1.3.0/docs/faq.md

@@ -0,0 +1,91 @@
+# FAQ
+
+## What is Agent OS?
+Agent OS is a kernel-style architecture for governing autonomous AI agents.  
+It applies operating system concepts—such as policies, signals, and controlled execution—to intercept and validate agent actions deterministically before they execute.
+
+Agent OS focuses on *governance*, not agent creation.
+
+---
+
+## How is Agent OS different from LangChain or CrewAI?
+LangChain and CrewAI are **agent frameworks** used to build agents.  
+Agent OS is **governance infrastructure** that wraps those frameworks.
+
+You can use Agent OS *alongside* LangChain, CrewAI, Semantic Kernel, or OpenAI Assistants to enforce safety, policies, and auditability during execution.
+
+---
+
+## Do I need to change my existing agent code?
+No.  
+Agent OS provides adapters that wrap existing agent code with minimal changes.
+
+Most integrations involve wrapping your current agent or executor with a KernelSpace instance and defining policies.
+
+---
+
+## What LLM providers are supported?
+Agent OS is model-agnostic.
+
+It works with:
+- OpenAI
+- Anthropic
+- Local models (Ollama, llama.cpp)
+- Any provider supported by LangChain, CrewAI, or Semantic Kernel
+
+Agent OS governs *actions*, not model outputs.
+
+---
+
+## How do I write custom policies?
+Policies are defined declaratively using rule-based configurations or helper APIs.
+
+Policies match known action patterns (e.g., file writes, SQL operations, API calls) and block or allow execution deterministically.
+
+Custom policies can be added by extending the policy engine or writing new rule definitions.
+
+---
+
+## Is Agent OS production-ready?
+The **core kernel components** (policy engine, signal handling, audit logging) are production-ready for controlled environments.
+
+Advanced modules—such as multi-agent trust, cross-model verification, and observability—are experimental and evolving.
+
+For strong isolation guarantees, Agent OS should be paired with container or sandbox-level isolation.
+
+---
+
+## What does “deterministic enforcement” mean?
+Deterministic enforcement means actions are blocked or allowed by the policy engine, not by prompting or relying on LLM behavior.
+
+If a rule matches an action, the action is stopped immediately—without asking the model to comply.
+
+---
+
+## Does Agent OS replace safety tools like Guardrails or LlamaGuard?
+No.  
+Those tools perform input/output filtering.
+
+Agent OS operates *during execution* by intercepting actions.  
+They can be used together for defense in depth.
+
+---
+
+## Can Agent OS be used with local models?
+Yes.  
+Agent OS does not depend on cloud APIs and works with local inference stacks.
+
+Governance logic runs at the application layer, independent of the model runtime.
+
+---
+
+## How can I contribute?
+Contributions are welcome.
+
+You can:
+- Improve documentation
+- Add integrations or examples
+- Propose RFCs
+- Fix bugs or improve tests
+
+See the Contributing Guide and GitHub issues labeled `good first issue`.

{agent_os_kernel-1.2.0 → agent_os_kernel-1.3.0}/docs/index.md

@@ -42,6 +42,7 @@ Learn by doing with our Jupyter notebooks:
 - [Framework Integrations](integrations.md) - LangChain, OpenAI, CrewAI
 - [Dependencies](dependencies.md) - Package dependencies
 - [Security Specification](security-spec.md) - Security model
+- [FAQ](faq.md) - Common questions and answers
 
 ### 📋 RFCs
 

{agent_os_kernel-1.2.0 → agent_os_kernel-1.3.0}/modules/amb/amb_core/__init__.py

@@ -109,6 +109,20 @@ from amb_core.tracing import (
     extract_trace,
 )
 
+# CloudEvents support (AMB-008)
+from amb_core.cloudevents import (
+    CloudEvent,
+    CloudEventBatch,
+    to_cloudevent,
+    from_cloudevent,
+    to_http_headers,
+    from_http_headers,
+    topic_to_type,
+    type_to_topic,
+    CLOUDEVENTS_SPEC_VERSION,
+    CLOUDEVENTS_CONTENT_TYPE,
+)
+
 __all__ = [
     # Version info
     "__version__",
@@ -147,4 +161,15 @@ __all__ = [
     "get_current_trace",
     "inject_trace",
     "extract_trace",
+    # CloudEvents
+    "CloudEvent",
+    "CloudEventBatch",
+    "to_cloudevent",
+    "from_cloudevent",
+    "to_http_headers",
+    "from_http_headers",
+    "topic_to_type",
+    "type_to_topic",
+    "CLOUDEVENTS_SPEC_VERSION",
+    "CLOUDEVENTS_CONTENT_TYPE",
 ]