io.github.imran-siddique/agentos

编码与调试

by imran-siddique

Build and manage policy-compliant AI agents with safety enforcement and compliance checking

什么是 io.github.imran-siddique/agentos

Build and manage policy-compliant AI agents with safety enforcement and compliance checking

README

<div align="center">

Agent OS — Community Edition

A kernel architecture for governing autonomous AI agents

GitHub Stars Sponsor License Python CI PyPI Downloads OWASP VS Code Extension Documentation Featured in awesome-llm-apps awesome-mcp-servers Integrated in agent-lightning Integrated in LlamaIndex awesome-AI-Agents awesome-agent-skills awesome-devops-mcp Downloads Discussions

If this project helps you, please star it! It helps others discover Agent OS.

📦 Install the full stack: pip install ai-agent-governance[full]PyPI | GitHub

Quick StartDocumentationVS Code ExtensionExamplesAgent HypervisorAgentMeshAgent SRE

<br/>

Open in Gitpod

Try Agent OS instantly in your browser - no installation required

</div>

Integrated With

<p align="center"> <a href="https://github.com/langgenius/dify-plugins/pull/2060"><img src="https://img.shields.io/badge/Dify-Merged-success?style=flat-square" alt="Dify"></a> <a href="https://github.com/run-llama/llama_index/pull/20644"><img src="https://img.shields.io/badge/LlamaIndex-Merged-success?style=flat-square" alt="LlamaIndex"></a> <a href="https://github.com/microsoft/agent-lightning/pull/478"><img src="https://img.shields.io/badge/Agent--Lightning-Merged-success?style=flat-square" alt="Agent-Lightning"></a> <a href="https://pypi.org/project/langgraph-trust/"><img src="https://img.shields.io/badge/LangGraph-PyPI-blue?style=flat-square" alt="LangGraph"></a> <a href="https://pypi.org/project/openai-agents-trust/"><img src="https://img.shields.io/badge/OpenAI_Agents-PyPI-blue?style=flat-square" alt="OpenAI Agents"></a> <a href="https://clawhub.ai/imran-siddique/agentmesh-governance"><img src="https://img.shields.io/badge/OpenClaw-ClawHub-purple?style=flat-square" alt="OpenClaw"></a> </p>

Featured In

<p align="center"> <a href="https://github.com/Shubhamsaboo/awesome-llm-apps"><img src="https://img.shields.io/badge/awesome--llm--apps-92K_%E2%AD%90-orange?style=flat-square" alt="awesome-llm-apps"></a> <a href="https://github.com/Jenqyang/Awesome-AI-Agents/pull/45"><img src="https://img.shields.io/badge/Awesome--AI--Agents-listed-orange?style=flat-square" alt="Awesome-AI-Agents"></a> <a href="https://github.com/github/awesome-copilot/pull/755"><img src="https://img.shields.io/badge/awesome--copilot-listed-orange?style=flat-square" alt="awesome-copilot"></a> <a href="https://github.com/heilcheng/awesome-agent-skills/pull/34"><img src="https://img.shields.io/badge/awesome--agent--skills-listed-orange?style=flat-square" alt="awesome-agent-skills"></a> <a href="https://github.com/TensorBlock/awesome-mcp-servers/pull/66"><img src="https://img.shields.io/badge/awesome--mcp--servers-listed-orange?style=flat-square" alt="awesome-mcp-servers"></a> <a href="https://github.com/rohitg00/awesome-devops-mcp-servers/pull/27"><img src="https://img.shields.io/badge/awesome--devops--mcp-listed-orange?style=flat-square" alt="awesome-devops-mcp"></a> </p>across 170K+ GitHub stars. Governance for [Dify](https://github.com/langgenius/dify-plugins/pull/2060) (65K ⭐), [LlamaIndex](https://github.com/run-llama/llama_index/pull/20644) (47K ⭐), [LangGraph](https://pypi.org/project/langgraph-trust/), [OpenAI Agents](https://pypi.org/project/openai-agents-trust/), [Agent-Lightning](https://github.com/microsoft/agent-lightning/pull/478), and [OpenClaw](https://clawhub.ai/imran-siddique/agentmesh-governance).

📊 By The Numbers

<table> <tr> <td align="center"><h3>1,680+</h3><sub>Tests Passing</sub></td> <td align="center"><h3>12</h3><sub>Framework Integrations</sub></td> <td align="center"><h3>170K+</h3><sub>Combined Stars of<br/>Integrated Projects</sub></td> <td align="center"><h3>&lt;0.1ms p99</h3><sub>Governance Latency<br/><a href="benchmarks/results/BENCHMARKS.md">Benchmarks</a></sub></td> <td align="center"><h3>9</h3><sub>More Framework Proposals<br/>Under Review</sub></td> </tr> </table>

🏢 Adopted By Leading AI Frameworks

FrameworkStarsStatusLink
Dify65K ⭐✅ Mergeddify-plugins#2060
LlamaIndex47K ⭐✅ Mergedllama_index#20644
Microsoft Agent-Lightning15K ⭐✅ Mergedagent-lightning#478
LangGraph24K ⭐📦 Published on PyPIlanggraph-trust
OpenAI Agents SDK📦 Published on PyPIopenai-agents-trust
OpenClaw📦 Published on ClawHubagentmesh-governance
<details> <summary><b>📋 Proposals under review at 10+ frameworks (click to expand)</b></summary>
FrameworkStarsProposal
AutoGen54K ⭐microsoft/autogen#7242
CrewAI44K ⭐crewAIInc/crewAI#4502
Haystack22K ⭐deepset-ai/haystack#10615
Semantic Kernel27K ⭐microsoft/semantic-kernel#13556
smolagents25K ⭐✅ Adapter built — huggingface/smolagents#1989
LangGraph24K ⭐langchain-ai/langgraph#6824
Google ADK18K ⭐✅ Adapter built — google/adk-python#4517
PydanticAI15K ⭐✅ Adapter built — pydantic/pydantic-ai#4335
OpenAI Agents SDKopenai/openai-agents-python#2515
A2A Protocol21K ⭐a2aproject/A2A#1501
Oracle Agent Specoracle/agent-spec#105
AI Card Specagent-card/ai-card#16
</details>

💡 Why Agent OS?

The AI agent market is projected to reach $47B by 2030. As enterprises deploy autonomous agents at scale, governance becomes the critical infrastructure layer. Agent OS is the kernel that ensures every agent action is policy-enforced, auditable, and compliant — making AI agents enterprise-ready.

The problem: AI agents can execute arbitrary tools, access sensitive data, and make autonomous decisions — with no built-in governance, audit trails, or policy enforcement.

Our solution: A governance kernel that sits between agents and their actions, providing deterministic policy enforcement in <1ms with zero agent code changes.

How Agent OS Compares

ToolFocusWhen it acts
LangChain/CrewAIBuilding agentsN/A (framework)
NeMo GuardrailsInput/output filteringBefore/after LLM call
LlamaGuardContent classificationBefore/after LLM call
Agent OSAction interceptionDuring execution

Agent frameworks build agents. Safety tools filter I/O. Agent OS intercepts actions mid-execution — the only kernel-level governance layer.

🛡️ OWASP Agentic Top 10 Coverage

Agent OS + ecosystem covers 8 out of 10 OWASP Agentic Application Security risks:

RiskCoverageModule
ASI01 Agent Goal Hijack✅ FullGovernancePolicy.blocked_patterns
ASI02 Tool Misuse✅ FullMCPGateway — tool filtering, rate limiting, audit
ASI03 Identity & Privilege✅ Fullrequire_human_approval, RBAC policies
ASI04 Supply Chain⚠️ PartialTool allowlisting (no deep scanning yet)
ASI05 Code Execution✅ Fullblocked_patterns, sandbox integration
ASI06 Memory Poisoning✅ FullMemoryGuard — hash integrity, injection detection
ASI07 Inter-Agent Comms✅ FullAgentMesh trust handshake, HMAC auth
ASI08 Cascading Failures✅ FullAgent SRE circuit breakers, cascade detection
ASI09 Human-Agent Trust✅ FullHuman approval workflows, audit logging
ASI10 Rogue Agents⚠️ PartialAgent Hypervisor execution rings, kill switch

📄 Full OWASP mapping →

🌐 The Agent Governance Ecosystem

LayerPackagePurposeInstall
KernelAgent OSPolicy enforcement, action interceptionpip install agent-os-kernel
NetworkAgentMeshIdentity, trust, delegationpip install agentmesh-platform
ReliabilityAgent SRESLOs, chaos testing, circuit breakerspip install agent-sre
RuntimeAgent HypervisorExecution rings, resource limits, sagapip install agent-hypervisor
Full Stackai-agent-governanceAll of the abovepip install ai-agent-governance[full]

⚡ Quick Start in 30 Seconds

bash
pip install agent-os-kernel
python
from agent_os import StatelessKernel, ExecutionContext

# Create a governed agent in 3 lines
kernel = StatelessKernel()

# Define execution context with governance policies
ctx = ExecutionContext(agent_id="demo-agent", policies=["read_only"])

# Your agent runs with policy enforcement
result = await kernel.execute(
    action="database_query",
    params={"query": "SELECT * FROM users"},
    context=ctx
)

# ✅ Safe queries execute
# ❌ "DROP TABLE users" → Blocked by kernel

That's it! Your agent now has deterministic policy enforcement. Learn more →

🎬 See all features in action:

bash
git clone https://github.com/imran-siddique/agent-os && python agent-os/demo.py
<details> <summary><b>📋 More examples (click to expand)</b></summary>

Policy enforcement with custom rules

python
from agent_os import StatelessKernel

kernel = StatelessKernel()
kernel.load_policy_yaml("""
version: "1.0"
name: api-safety
rules:
  - name: block-destructive-sql
    condition: "action == 'database_query'"
    action: deny
    pattern: "DROP|TRUNCATE|DELETE FROM .* WHERE 1=1"
  - name: rate-limit-api
    condition: "action == 'api_call'"
    limit: "100/hour"
""")

result = await kernel.execute(action="database_query", params={"query": "DROP TABLE users"})
# ❌ Blocked: Matched rule 'block-destructive-sql'

Audit logging

python
from agent_os import KernelSpace

kernel = KernelSpace()

# Every kernel action is automatically recorded
result = await kernel.execute(action="read_file", params={"path": "/data/report.csv"})

# Query the flight recorder
entries = kernel.flight_recorder.query(agent_id="agent-001", limit=10)
for entry in entries:
    print(f"{entry.timestamp} | {entry.action} | {entry.outcome}")

Governed chatbot with memory

python
from agent_os import KernelSpace
from agent_os.emk import EpisodicMemory

kernel = KernelSpace(policy_file="policies.yaml")
memory = EpisodicMemory(max_turns=50)

@kernel.register
async def chat(message: str, conversation_id: str = "default") -> str:
    history = memory.get_history(conversation_id)
    response = await call_llm(history + [{"role": "user", "content": message}])
    memory.add_turn(conversation_id, message, response)
    return response
# Outputs are checked against content policies; violations trigger SIGSTOP

See examples/ for 20+ runnable demos including SQL agents, GitHub reviewers, and compliance bots.

</details>
<p align="center"> <img src="assets/demo-terminal.svg" alt="Agent OS Terminal Demo" width="700"/> </p>

🎯 What You'll Build in 5 Minutes

python
from agent_os import stateless_execute

# 1. Define safety policies (not prompts — actual enforcement)

# 2. Actions are checked against policies before execution
result = await stateless_execute(
    action="database_query",
    params={"query": "SELECT revenue FROM sales"},
    agent_id="analyst-001",
    policies=["read_only"]
)
# ✅ Safe queries execute
# ❌ "DROP TABLE users" → BLOCKED (not by prompt, by kernel)

Result: Defined policies are deterministically enforced by the kernel—not by hoping the LLM follows instructions.

For the full kernel with signals, VFS, and protection rings:

python
from agent_os import KernelSpace, AgentSignal, AgentVFS

# Requires: pip install agent-os-kernel[full]
kernel = KernelSpace()
ctx = kernel.create_agent_context("agent-001")
await ctx.write("/mem/working/task.txt", "Hello World")

Note: KernelSpace, AgentSignal, and AgentVFS require installing the control-plane module: pip install agent-os-kernel[full]


What is Agent OS?

Agent OS applies operating system concepts to AI agent governance. Instead of relying on prompts to enforce safety ("please don't do dangerous things"), it provides application-level middleware that intercepts and validates agent actions before execution.

Note: This is application-level enforcement (Python middleware), not OS kernel-level isolation. Agents run in the same process. For true isolation, run agents in containers.

code
┌─────────────────────────────────────────────────────────┐
│              USER SPACE (Agent Code)                    │
│   Your agent code runs here. The kernel intercepts      │
│   actions before they execute.                          │
├─────────────────────────────────────────────────────────┤
│              KERNEL SPACE                               │
│   Policy Engine │ Flight Recorder │ Signal Dispatch     │
│   Actions are checked against policies before execution │
└─────────────────────────────────────────────────────────┘

The Idea

Prompt-based safety asks the LLM to follow rules. The LLM decides whether to comply.

Kernel-based safety intercepts actions before execution. The policy engine decides, not the LLM.

This is the same principle operating systems use: applications request resources, the kernel grants or denies access based on permissions.


Architecture

mermaid
graph TB
    subgraph Frameworks["External Frameworks"]
        LC[LangChain]
        CA[CrewAI]
        AG[AutoGen]
        OA[OpenAI]
        AN[Anthropic]
        GE[Gemini]
    end

    ADAPT[[Adapters]]

    subgraph L4["Layer 4: Intelligence"]
        SC[Self-Correction]
        SA[Semantic Analysis]
    end

    subgraph L3["Layer 3: Control Plane"]
        PE[Policy Engine]
        AL[Audit Logging]
    end

    subgraph L2["Layer 2: Infrastructure"]
        NM[Nexus Marketplace]
        OB[Observability]
    end

    subgraph L1["Layer 1: Primitives"]
        VE[Verification Engine]
        CS[Context Service]
        MS[Memory Store]
    end

    LC & CA & AG & OA & AN & GE --> ADAPT
    ADAPT --> L4
    SC & SA --> PE
    SC & SA --> AL
    PE & AL --> NM
    PE & AL --> OB
    NM & OB --> VE
    NM & OB --> CS
    NM & OB --> MS

Directory Structure

code
agent-os/
├── src/agent_os/             # Core Python package
│   ├── __init__.py           # Public API (re-exports from all layers)
│   ├── stateless.py          # StatelessKernel (zero-dependency core)
│   ├── base_agent.py         # BaseAgent, ToolUsingAgent classes
│   ├── agents_compat.py      # AGENTS.md parser (OpenAI/Anthropic standard)
│   ├── cli.py                # CLI (agent-os check, review, init, etc.)
│   └── integrations/         # Framework adapters (LangChain, OpenAI, etc.)
├── modules/                  # Kernel Modules (4-layer architecture)
│   ├── primitives/           # Layer 1: Base types and failures
│   ├── cmvk/                 # Layer 1: Verification
│   ├── emk/                  # Layer 1: Episodic memory kernel
│   ├── caas/                 # Layer 1: Context-as-a-Service
│   ├── amb/                  # Layer 2: Agent message bus
│   ├── iatp/                 # Layer 2: Inter-agent trust protocol
│   ├── atr/                  # Layer 2: Agent tool registry
│   ├── observability/        # Layer 3: Prometheus + OpenTelemetry
│   ├── control-plane/        # Layer 3: THE KERNEL (policies, signals)
│   ├── scak/                 # Layer 4: Self-correcting agent kernel
│   ├── mute-agent/           # Layer 4: Face/Hands architecture
│   ├── nexus/                # Experimental: Trust exchange network
│   └── mcp-kernel-server/    # Integration: MCP protocol support
├── extensions/               # IDE & AI Assistant Extensions
│   ├── mcp-server/           # ⭐ MCP Server (Copilot, Claude, Cursor)
│   ├── vscode/               # VS Code extension
│   ├── copilot/              # GitHub Copilot extension
│   ├── jetbrains/            # IntelliJ/PyCharm plugin
│   ├── cursor/               # Cursor IDE extension
│   ├── chrome/               # Chrome extension
│   └── github-cli/           # gh CLI extension
├── examples/                 # Working examples
├── docs/                     # Documentation
├── tests/                    # Test suite (organized by layer)
├── notebooks/                # Jupyter tutorials
├── papers/                   # Research papers
└── templates/                # Policy templates

Core Modules

ModuleLayerPyPI PackageDescriptionStatus
primitives1agent-primitivesBase failure types, severity levels✅ Stable
cmvk1cmvkVerification, drift detection✅ Stable
emk1emkEpisodic memory kernel (append-only ledger)✅ Stable
caas1caas-coreContext-as-a-Service, RAG pipeline✅ Stable
amb2amb-coreAgent message bus (async pub/sub)✅ Stable
iatp2inter-agent-trust-protocolSidecar trust protocol, typed IPC pipes✅ Stable
atr2agent-tool-registryTool registry with LLM schema generation✅ Stable
control-plane3agent-control-planeTHE KERNEL — Policy engine, signals, VFS✅ Stable
observability3agent-os-observabilityPrometheus metrics + OpenTelemetry tracing⚠️ No tests
scak4scakSelf-correcting agent kernel✅ Stable
mute-agent4mute-agentDecoupled reasoning/execution architecture⚠️ No tests
nexusNot publishedTrust exchange network🔬 Prototype
mcp-kernel-serverIntmcp-kernel-serverMCP server for Claude Desktop⚠️ No tests
hypervisoragent-hypervisorRuntime supervisor — Execution Rings, Joint Liability, Saga Orchestrator (own repo)✅ 184 tests

⭐ Star Feature: Agent Hypervisor

Runtime supervisor for multi-agent collaboration — think "VMware for AI agents."

Now its own repo: agent-hypervisor — 184 tests, 268μs full pipeline, zero dependencies beyond pydantic.

Just as OS hypervisors isolate virtual machines and enforce resource boundaries, the Agent Hypervisor isolates AI agent sessions and enforces governance boundaries at sub-millisecond latency.

code
┌────────────────────────────────────────────────────────────┐
│                    AGENT HYPERVISOR                         │
│                                                            │
│   Ring 0 (Root)      ← SRE Witness required                │
│   Ring 1 (Privileged)← σ_eff > 0.95 + consensus           │
│   Ring 2 (Standard)  ← σ_eff > 0.60                        │
│   Ring 3 (Sandbox)   ← Default for unknown agents          │
│                                                            │
│   ┌──────────┐  ┌───────────┐  ┌────────────────────────┐  │
│   │  Joint    │  │  Semantic  │  │  Hash-Chained          │  │
│   │ Liability │  │   Saga     │  │  Delta Audit Trail     │  │
│   │  Engine   │  │ Orchestr.  │  │  (Tamper-Evident)      │  │
│   └──────────┘  └───────────┘  └────────────────────────┘  │
└────────────────────────────────────────────────────────────┘

Key Capabilities

FeatureDescriptionLatency
Execution Rings4-level privilege model (Ring 0–3) based on trust score0.3μs
Joint LiabilityHigh-trust agents vouch for low-trust agents with bonded reputation7μs
Saga OrchestratorMulti-step transactions with timeout, retry, and auto-compensation151μs
Delta AuditHash-chained semantic diffs with blockchain commitment27μs
Full PipelineSession + join + audit + saga + terminate268μs

Quick Start

bash
pip install agent-hypervisor
python
from hypervisor import Hypervisor, SessionConfig, ConsistencyMode

hv = Hypervisor()

# Create a governed multi-agent session
session = await hv.create_session(
    config=SessionConfig(consistency_mode=ConsistencyMode.EVENTUAL, max_participants=5),
    creator_did="did:mesh:admin",
)

# Agents are automatically assigned privilege rings based on trust score
ring = await hv.join_session(session.sso.session_id, "did:mesh:agent-alpha", sigma_raw=0.85)
# → Ring 2 (Standard) — can execute reversible actions

# Multi-step saga with automatic timeout and compensation
saga = session.saga.create_saga(session.sso.session_id)
step = session.saga.add_step(
    saga.saga_id, "draft_email", "did:mesh:agent-alpha",
    execute_api="/api/draft", undo_api="/api/undo-draft",
    timeout_seconds=30, max_retries=2,
)

# Terminate — returns tamper-evident summary hash
summary_hash = await hv.terminate_session(session.sso.session_id)

📖 Full Hypervisor documentation →


IDE & CLI Extensions

ExtensionDescriptionStatus
mcp-serverMCP Server — Works with Claude, Copilot, Cursor (npx agentos-mcp-server)✅ Published (v1.0.1)
vscodeVS Code extension with real-time policy checks, enterprise features✅ Published (v1.0.1)
copilotGitHub Copilot extension (Vercel/Docker deployment)✅ Published (v1.0.0)
jetbrainsIntelliJ, PyCharm, WebStorm plugin (Kotlin)✅ Built (v1.0.0)
cursorCursor IDE extension (Composer integration)✅ Built (v0.1.0)
chromeChrome extension for GitHub, Jira, AWS, GitLab✅ Built (v1.0.0)
github-cligh agent-os CLI extension⚠️ Basic

Install

bash
pip install agent-os-kernel

Or with optional components:

bash
pip install agent-os-kernel[cmvk]           # + verification
pip install agent-os-kernel[iatp]           # + inter-agent trust
pip install agent-os-kernel[observability]  # + Prometheus/OpenTelemetry
pip install agent-os-kernel[nexus]          # + trust exchange network
pip install agent-os-kernel[full]           # Everything

One-Command Quickstart

macOS/Linux:

bash
curl -sSL https://raw.githubusercontent.com/imran-siddique/agent-os/main/scripts/quickstart.sh | bash

Windows (PowerShell):

powershell
iwr -useb https://raw.githubusercontent.com/imran-siddique/agent-os/main/scripts/quickstart.ps1 | iex

Quick Example

Stateless API (Always Available — Zero Dependencies Beyond Pydantic)

python
from agent_os import stateless_execute

# Execute with policy enforcement
result = await stateless_execute(
    action="database_query",
    params={"query": "SELECT * FROM users"},
    agent_id="analyst-001",
    policies=["read_only"]
)

Full Kernel API (Requires pip install agent-os-kernel[full])

python
from agent_os import KernelSpace, AgentSignal, PolicyRule

kernel = KernelSpace()

# Create agent context with VFS
ctx = kernel.create_agent_context("agent-001")
await ctx.write("/mem/working/task.txt", "analyze this data")

# Policy enforcement
from agent_os import PolicyEngine
engine = PolicyEngine()
engine.add_rule(PolicyRule(name="no_sql_injection", pattern="DROP|DELETE|TRUNCATE"))

POSIX-Inspired Primitives

Agent OS borrows concepts from POSIX operating systems:

ConceptPOSIXAgent OS
Process controlSIGKILL, SIGSTOPAgentSignal.SIGKILL, AgentSignal.SIGSTOP
Filesystem/proc, /tmpVFS with /mem/working, /mem/episodic
IPCPipes (|)Typed IPC pipes between agents
Syscallsopen(), read()kernel.execute()

Signals

python
# Requires: pip install agent-os-kernel[full]
from agent_os import SignalDispatcher, AgentSignal

dispatcher = SignalDispatcher()
dispatcher.signal(agent_id, AgentSignal.SIGSTOP)  # Pause
dispatcher.signal(agent_id, AgentSignal.SIGCONT)  # Resume
dispatcher.signal(agent_id, AgentSignal.SIGKILL)  # Terminate

VFS (Virtual File System)

python
# Requires: pip install agent-os-kernel[full]
from agent_os import AgentVFS

vfs = AgentVFS(agent_id="agent-001")
vfs.write("/mem/working/task.txt", "Current task")
vfs.read("/policy/rules.yaml")  # Read-only from user space

Framework Integrations

Wrap existing frameworks with Agent OS governance:

python
# LangChain
from agent_os.integrations import LangChainKernel
governed = LangChainKernel().wrap(my_chain)

# OpenAI Assistants
from agent_os.integrations import OpenAIKernel
governed = OpenAIKernel().wrap_assistant(assistant, client)

# Semantic Kernel
from agent_os.integrations import SemanticKernelWrapper
governed = SemanticKernelWrapper().wrap(sk_kernel)

# CrewAI
from agent_os.integrations import CrewAIKernel
governed = CrewAIKernel().wrap(my_crew)

# AutoGen
from agent_os.integrations import AutoGenKernel
governed = AutoGenKernel().wrap(autogen_agent)

# OpenAI Agents SDK
from agent_os.integrations import OpenAIAgentsSDKKernel
governed = OpenAIAgentsSDKKernel().wrap(agent)

Note: These adapters use lazy interception — they don't require the target framework to be installed until you call .wrap().

See integrations documentation for full details.

Integration Comparison

FrameworkGovernance LevelAsync SupportStatusAdapter File
LangChainChain/Agent/Runnableainvoke✅ Stableintegrations/langchain_adapter.py
OpenAI AssistantsRun/Thread/Tool Call✅ Streaming✅ Stableintegrations/openai_adapter.py
AutoGenMulti-Agent Orchestration❌ Sync only✅ Stableintegrations/autogen_adapter.py
Semantic KernelFunction/Plugin/Memory✅ Native async✅ Stableintegrations/semantic_kernel_adapter.py
CrewAICrew/Agent/Task❌ Sync only✅ Stableintegrations/crewai_adapter.py
OpenAI Agents SDKAgent/Tool/Handoff✅ Native async✅ Stableintegrations/openai_agents_sdk_adapter.py

Examples

The examples/ directory contains demos at various levels:

Getting Started

DemoDescriptionCommand
demo-appUses the stateless API (most reliable)cd examples/demo-app && python demo.py
hello-worldMinimal examplecd examples/hello-world && python agent.py
quickstartQuick introcd examples/quickstart && python my_first_agent.py

Domain Examples (Self-Contained)

These examples are self-contained and don't require external Agent OS imports:

DemoDescription
healthcare-hipaaHIPAA-compliant agent
customer-serviceCustomer support agent
legal-reviewLegal document analysis
crewai-safe-modeCrewAI with safety wrappers

Production Demos (with Docker + Observability)

DemoDescriptionCommand
carbon-auditorMulti-model verificationcd examples/carbon-auditor && docker-compose up
grid-balancingMulti-agent coordinationcd examples/grid-balancing && docker-compose up
defi-sentinelReal-time attack detectioncd examples/defi-sentinel && docker-compose up
pharma-complianceDocument analysiscd examples/pharma-compliance && docker-compose up

Each production demo includes:

  • Grafana dashboard on port 300X
  • Prometheus metrics on port 909X
  • Jaeger tracing on port 1668X
bash
# Run carbon auditor with full observability
cd examples/carbon-auditor
cp .env.example .env  # Optional: add API keys
docker-compose up

# Open dashboards
open http://localhost:3000  # Grafana (admin/admin)
open http://localhost:16686 # Jaeger traces

Safe Tool Plugins

Agent OS includes pre-built safe tools via the Agent Tool Registry:

python
# Requires: pip install agent-os-kernel[full]
from atr import ToolRegistry, tool

@tool(name="safe_http", description="Rate-limited HTTP requests")
async def safe_http(url: str) -> dict:
    # Tool is automatically registered and sandboxed
    ...

registry = ToolRegistry()
registry.register(safe_http)

# Generate schemas for any LLM
openai_tools = registry.to_openai_schema()
anthropic_tools = registry.to_anthropic_schema()

Message Bus

Connect agents using the async message bus:

python
# Requires: pip install agent-os-kernel[full]
from amb_core import MessageBus, Message

bus = MessageBus()
await bus.subscribe("tasks", handler)
await bus.publish("tasks", Message(payload={"task": "analyze"}))

Broker adapters available for Redis, Kafka, and NATS (requires optional dependencies).


CLI Tool

Agent OS includes a CLI for terminal workflows:

bash
# Check files for safety violations
agentos check src/app.py
# ✓ src/app.py: No violations
# OR
# ⚠️  2 violation(s) found in src/app.py:
#   Line 12: DROP TABLE users;
#     Violation: Destructive SQL: DROP operation detected
#     Policy: block-destructive-sql

# Check staged git files (ideal for pre-commit hooks)
agentos check --staged
# ✓ No violations in staged files

# Machine-readable JSON output (for CI pipelines)
agentos check src/app.py --format json

# CI mode (no colors, strict exit codes)
agentos check --staged --ci
bash
# Initialize Agent OS in a project
agentos init
# Initialized Agent OS in .agents/
#   - agents.md: Agent instructions (OpenAI/Anthropic standard)
#   - security.md: Kernel policies (Agent OS extension)
#   - Template: strict

# Choose a permissive or audit-only template
agentos init --template permissive
agentos init --template audit

# Overwrite an existing .agents/ directory
agentos init --force
bash
# Enable kernel governance and verify the configuration
agentos secure
# Securing agents in .
#   [PASS] kernel version
#   [PASS] signals defined
#   [PASS] policies defined
# Security configuration valid.
bash
# Audit agent security configuration
agentos audit
# Auditing .
#   [OK] agents.md
#   [OK] security.md
# No issues found.

# JSON output for CI
agentos audit --format json
bash
# Show kernel status (version, installed packages)
agentos status
# Agent OS Kernel Status
# ========================================
#   Version: 1.2.0
#   Status: Installed
#   Project: /home/user/myproject
#   Agents: Configured (.agents/ found)
bash
# Multi-model code review with CMVK consensus
agentos review src/app.py --cmvk
# 🔍 Reviewing src/app.py with CMVK...
# Multi-Model Review (3 models):
#   ✅ gpt-4: No issues
#   ⚠️  claude-sonnet-4: 1 potential issue(s)
#   ✅ gemini-pro: No issues
# Consensus: 67%

# Specify models
agentos review src/app.py --cmvk --models "gpt-4,claude-sonnet-4"
bash
# Validate policy YAML files
agentos validate
#   Checking .agents/policy.yaml... OK
# ✓ All 1 policy file(s) valid.

# Validate specific files in strict mode
agentos validate policies/*.yaml --strict
bash
# Install git pre-commit hook
agentos install-hooks
# ✓ Installed pre-commit hook: .git/hooks/pre-commit
# Agent OS will now check staged files before each commit.

# Append to an existing hook
agentos install-hooks --append
bash
# Start the HTTP API server
agentos serve --port 8080
# Agent OS API server starting on 0.0.0.0:8080
# Endpoints:
#   GET  /health              Health check
#   GET  /status              Kernel status
#   GET  /agents              List agents
#   POST /agents/{id}/execute  Execute agent action
bash
# Output Prometheus-style metrics
agentos metrics
# # HELP agentos_policy_violations_total Total policy violations.
# # TYPE agentos_policy_violations_total counter
# agentos_policy_violations_total 0
# ...

MCP Integration (Claude Desktop, GitHub Copilot, Cursor)

Agent OS provides an MCP server that works with any MCP-compatible AI assistant:

bash
# Quick install via npx
npx agentos-mcp-server

npm: agentos-mcp-server
MCP Registry: io.github.imran-siddique/agentos

Add to your config file:

Claude Desktop (%APPDATA%\Claude\claude_desktop_config.json on Windows):

json
{
  "mcpServers": {
    "agentos": {
      "command": "npx",
      "args": ["-y", "agentos-mcp-server"]
    }
  }
}

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 for full details.


Documentation

Tutorials

Interactive Notebooks

NotebookDescriptionTime
Hello Agent OSYour first governed agent5 min
Episodic MemoryAgent memory that persists15 min
Time-Travel DebuggingReplay and debug decisions20 min
VerificationDetect hallucinations15 min
Multi-Agent CoordinationTrust between agents20 min
Policy EngineDeep dive into policies15 min

Reference


Status & Maturity

This is a research project exploring kernel concepts for AI agent governance.

✅ Production-Ready

These components are fully implemented and tested:

ComponentTests
StatelessKernel — Zero-dependency policy enforcement (src/agent_os/)✅ Full coverage
Policy Engine — Deterministic rule enforcement✅ Tested
Flight Recorder — SQLite-based audit logging✅ Tested
CLIagent-os check, init, secure, validate✅ Tested
Framework Adapters — LangChain, OpenAI, Semantic Kernel, CrewAI, AutoGen, OpenAI Agents SDK✅ Implemented
AGENTS.md Parser — OpenAI/Anthropic standard agent config✅ Full coverage
Primitives (agent-primitives) — Failure types, severity levels✅ Tested
CMVK (cmvk) — Drift detection, distance metrics (955+ lines)✅ Tested
EMK (emk) — Episodic memory with JSONL storage✅ 8 test files
AMB (amb-core) — Async message bus, DLQ, tracing✅ 6 test files
IATP (inter-agent-trust-protocol) — Sidecar trust, typed IPC✅ 9 test files
ATR (agent-tool-registry) — Multi-LLM schema generation✅ 6 test files
Control Plane (agent-control-plane) — Signals, VFS, protection rings✅ 18 test files
SCAK (scak) — Self-correcting agent kernel✅ 23 test files

⚠️ Experimental (Code Exists, Tests Missing or Incomplete)

ComponentWhat's Missing
Mute Agent (mute-agent)No tests; all layer dependencies use mock adapters
Observability (agent-os-observability)No tests; Prometheus metrics, Grafana dashboards, OTel tracing implemented
MCP Kernel Server (mcp-kernel-server)No tests; 1173-line implementation
GitHub CLI ExtensionSingle bash script with simulated output
Control Plane MCP AdapterPlaceholder — returns canned responses
Control Plane A2A AdapterPlaceholder — negotiation accepts all params

🔬 Research Prototype

ComponentWhat's Missing
Nexus Trust ExchangeNo pyproject.toml, no tests, placeholder cryptography (XOR — not secure), all signature verification stubbed, in-memory storage only

Known Architectural Limitations

LimitationImpactMitigation
Application-level onlyDirect stdlib calls (subprocess, open) bypass kernelPair with container isolation for production
Blocklist-based policiesNovel attack patterns not in rules will passAdd AST-level parsing (#32), use defense in depth
Shadow Mode single-stepMulti-step agent simulations diverge from realityUse for single-turn validation only
No tamper-proof auditFlight Recorder SQLite can be modified by compromised agentWrite to external sink for critical audits
Provider-coupled adaptersEach SDK needs separate adapterAbstract interface planned (#47)

See GitHub Issues for the full roadmap.


FAQ

How is this different from prompt-based safety?

Prompt-based safety relies on instructing the LLM to follow rules via system prompts. This approach is probabilistic — the model may still produce unsafe outputs under certain conditions.

Agent OS enforces policies at the middleware layer. Actions are intercepted and validated before execution, making enforcement deterministic rather than dependent on model compliance.

What frameworks are supported?

Agent OS can wrap and govern agents built with popular frameworks including LangChain, CrewAI, AutoGen, Semantic Kernel, and the OpenAI SDK. It also supports MCP-based integrations.

Can I use this in production?

Core components such as the StatelessKernel and Policy Engine are production-ready. However, Agent OS provides application-level enforcement. For high-security environments, it should be combined with infrastructure isolation (e.g., containers).

How do I write custom policies?

Custom policies can be defined programmatically in Python or declaratively using YAML. Policies define rules that inspect and allow or deny agent actions before execution.

What is the performance overhead?

Policy checks are lightweight and typically introduce only minimal latency per action. The overhead depends on the number and complexity of rules configured.


Troubleshooting

Common Issues

ModuleNotFoundError: No module named 'agent_os'

bash
# Install from source
git clone https://github.com/imran-siddique/agent-os.git
cd agent-os
pip install -e .

Optional modules not available

bash
# Check what's installed
python -c "from agent_os import check_installation; check_installation()"

# Install everything
pip install -e ".[full]"

Permission errors on Windows

bash
# Run PowerShell as Administrator, or use --user flag
pip install --user -e .

Docker not working

bash
# Build with Dockerfile (no Docker Compose needed for simple tests)
docker build -t agent-os .
docker run -it agent-os python examples/demo-app/demo.py

Tests failing with API errors

bash
# Most tests work without API keys — mock mode is default
pytest tests/ -v

# For real LLM tests, set environment variables
export OPENAI_API_KEY=sk-...
export ANTHROPIC_API_KEY=sk-ant-...

Frequently Asked Questions

What is the difference between Agent OS and prompt-based guardrails? Prompt-based guardrails ask the LLM to self-police, which is probabilistic. Agent OS enforces governance at the middleware level using deterministic policy engines and POSIX-inspired access controls. It controls what agents can do (capability-based), not just what they should not do (filter-based).

How does Agent OS work with other frameworks? Agent OS integrates with 14+ frameworks via adapters. Install the governance layer alongside your existing framework: use langgraph-trust for LangGraph, openai-agents-trust for OpenAI Agents, or the MCP server for any MCP-compatible client. Agent OS acts as a kernel layer underneath your agent framework.

What is the Agent Governance Ecosystem? Agent OS is part of a suite of four projects: Agent OS (policy kernel), AgentMesh (trust network), Agent Hypervisor (runtime supervisor), and Agent SRE (reliability platform). Together they provide 4,310+ tests across 17 modules.

Can I use Agent OS in production? Yes. Agent OS has 1,500+ tests, a VS Code extension, PyPI package (pip install agent-os-kernel), and is integrated into production frameworks like Dify (65K stars) and LlamaIndex (47K stars). It supports Python 3.9+ and runs on any platform.


Contributing

bash
git clone https://github.com/imran-siddique/agent-os.git
cd agent-os
pip install -e ".[dev]"
pytest

License

MIT — See LICENSE


<div align="center">

Exploring kernel concepts for AI agent safety.

GitHub · Docs

</div>

常见问题

io.github.imran-siddique/agentos 是什么?

Build and manage policy-compliant AI agents with safety enforcement and compliance checking

相关 Skills

前端设计

by anthropics

Universal
热门

面向组件、页面、海报和 Web 应用开发,按鲜明视觉方向生成可直接落地的前端代码与高质感 UI,适合做 landing page、Dashboard 或美化现有界面,避开千篇一律的 AI 审美。

想把页面做得既能上线又有设计感,就用前端设计:组件到整站都能产出,难得的是能避开千篇一律的 AI 味。

编码与调试
未扫描111.1k

网页构建器

by anthropics

Universal
热门

面向复杂 claude.ai HTML artifact 开发,快速初始化 React + Tailwind CSS + shadcn/ui 项目并打包为单文件 HTML,适合需要状态管理、路由或多组件交互的页面。

在 claude.ai 里做复杂网页 Artifact 很省心,多组件、状态和路由都能顺手搭起来,React、Tailwind 与 shadcn/ui 组合效率高、成品也更精致。

编码与调试
未扫描111.1k

网页应用测试

by anthropics

Universal
热门

用 Playwright 为本地 Web 应用编写自动化测试,支持启动开发服务器、校验前端交互、排查 UI 异常、抓取截图与浏览器日志,适合调试动态页面和回归验证。

借助 Playwright 一站式验证本地 Web 应用前端功能,调 UI 时还能同步查看日志和截图,定位问题更快。

编码与调试
未扫描111.1k

相关 MCP Server

GitHub

编辑精选

by GitHub

热门

GitHub 是 MCP 官方参考服务器,让 Claude 直接读写你的代码仓库和 Issues。

这个参考服务器解决了开发者想让 AI 安全访问 GitHub 数据的问题,适合需要自动化代码审查或 Issue 管理的团队。但注意它只是参考实现,生产环境得自己加固安全。

编码与调试
83.0k

by Context7

热门

Context7 是实时拉取最新文档和代码示例的智能助手,让你告别过时资料。

它能解决开发者查找文档时信息滞后的问题,特别适合快速上手新库或跟进更新。不过,依赖外部源可能导致偶尔的数据延迟,建议结合官方文档使用。

编码与调试
51.7k

by tldraw

热门

tldraw 是让 AI 助手直接在无限画布上绘图和协作的 MCP 服务器。

这解决了 AI 只能输出文本、无法视觉化协作的痛点——想象让 Claude 帮你画流程图或白板讨论。最适合需要快速原型设计或头脑风暴的开发者。不过,目前它只是个基础连接器,你得自己搭建画布应用才能发挥全部潜力。

编码与调试
46.2k

评论