cortexhub 0.1.10

CortexHub Python SDK - Policy-as-Code for AI Agents

CortexHub Python SDK

Runtime Governance for AI Agents - Policy enforcement, PII/secrets detection, complete audit trails with OpenTelemetry.

Installation

# Core SDK
pip install cortexhub

# With framework support (choose one or more)
pip install cortexhub[langgraph]      # LangGraph
pip install cortexhub[crewai]         # CrewAI
pip install cortexhub[openai-agents]  # OpenAI Agents SDK
pip install cortexhub[claude-agents]  # Claude Agent SDK

# All frameworks (for development)
pip install cortexhub[all]

Python support: 3.9–3.12. Python 3.13 is not supported because the spaCy en-core-web-sm model does not publish 3.13 wheels.

Note: The SDK ships with the spaCy en-core-web-sm model for Presidio PII detection, so it should not download the model at runtime.

Quick Start

from cortexhub import init, Framework

# Initialize CortexHub FIRST, before importing your framework
cortex = init(
    agent_id="customer_support_agent",
    framework=Framework.LANGGRAPH,  # or CREWAI, OPENAI_AGENTS, CLAUDE_AGENTS
)

# Now import and use your framework
from langgraph.prebuilt import create_react_agent

# Continue with your LangGraph setup...

Supported Frameworks

Framework	Enum Value	Install
LangGraph	Framework.LANGGRAPH	pip install cortexhub[langgraph]
CrewAI	Framework.CREWAI	pip install cortexhub[crewai]
OpenAI Agents	Framework.OPENAI_AGENTS	pip install cortexhub[openai-agents]
Claude Agents	Framework.CLAUDE_AGENTS	pip install cortexhub[claude-agents]

Tracing Coverage

All frameworks emit run.started and run.completed/run.failed for each run. Tool spans (tool.invoke) and model spans (llm.call) vary by SDK:

• LangGraph: tool calls via BaseTool.invoke, LLM calls via BaseChatModel.invoke/ainvoke
• CrewAI: tool calls via CrewStructuredTool.invoke/BaseTool.run, LLM calls via LiteLLM and BaseLLM.call/acall
• OpenAI Agents: tool calls via function_tool, LLM calls via OpenAIResponsesModel and OpenAIChatCompletionsModel
• Claude Agents: tool calls via @tool and built-in tool hooks; LLM calls run inside the Claude Code CLI and are not intercepted by the Python SDK

Configuration

# Required: API key
export CORTEXHUB_API_KEY=ch_live_...

Features

• Policy Enforcement - Cloud configuration, local evaluation
• PII Detection - 50+ entity types, configurable
• Secrets Detection - 30+ secret types
• Configurable Guardrails - Select specific PII/secret types to redact
• Custom Patterns - Add company-specific regex patterns
• OpenTelemetry - Industry-standard observability
• Framework Adapters - Automatic interception for all major frameworks
• Privacy Mode - Metadata-only by default, safe for production

Privacy Modes

# Production (default) - only metadata sent
cortex = init(agent_id="...", framework=..., privacy=True)
# Sends: tool names, arg schemas, PII types detected
# Never: raw values, prompts, responses

# Development - full data for testing policies  
cortex = init(agent_id="...", framework=..., privacy=False)
# Also sends: raw args, results, prompts (for policy testing)

Policy Enforcement

Policies are created in the CortexHub dashboard from detected risks. The SDK automatically fetches and enforces them:

from cortexhub.errors import PolicyViolationError, ApprovalRequiredError

# Policies are fetched automatically during init()
# If policies exist, enforcement mode is enabled

try:
    agent.run("Process a $10,000 refund")
except PolicyViolationError as e:
    print(f"Blocked by policy: {e.policy_name}")
    print(f"Reason: {e.reasoning}")
except ApprovalRequiredError as e:
    print(f"\n⏸️  APPROVAL REQUIRED")
    print(f"   Approval ID: {e.approval_id}")
    print(f"   Tool: {e.tool_name}")
    print(f"   Reason: {e.reason}")
    print(f"   Expires: {e.expires_at}")
    print(f"\n   Decision endpoint: {e.decision_endpoint}")
    print(f"   Configure a webhook to receive approval.decisioned event")

Guardrail Configuration

Guardrails detect PII and secrets in LLM prompts. Configure in the dashboard:

1. Select types to redact: Choose specific PII types (email, phone, etc.)
2. Add custom patterns: Regex for company-specific data (employee IDs, etc.)
3. Choose action: Redact, block, or monitor only

The SDK applies your configuration automatically:

# With guardrail policy active:
# Input prompt: "Contact [e-mail] about employee EMP-123456"
# After redaction: "Contact [REDACTED-EMAIL_ADDRESS] about employee [REDACTED-CUSTOM_EMPLOYEE_ID]"
# Only configured types are redacted

Important: Initialization Order

Always initialize CortexHub FIRST, before importing your framework:

# ✅ CORRECT
from cortexhub import init, Framework
cortex = init(agent_id="my_agent", framework=Framework.LANGGRAPH)

from langgraph.prebuilt import create_react_agent  # Import AFTER init

# ❌ WRONG
from langgraph.prebuilt import create_react_agent  # Framework imported first
from cortexhub import init, Framework
cortex = init(...)  # Too late!

This ensures:

1. CortexHub sets up OpenTelemetry before frameworks that also use it
2. Framework decorators/classes are properly wrapped

Architecture

Agent Decides → [CortexHub] → Agent Executes
                    │
              ┌─────┴─────┐
              │           │
         Policy      Guardrails
         Engine      (PII/Secrets)
              │           │
              └─────┬─────┘
                    │
              OpenTelemetry
               (to backend)

Development

cd python

# Install with all frameworks
uv sync --all-extras

# Run tests
uv run pytest

# Lint
uv run ruff check .

Links

• Documentation
• Dashboard
• Issues

License

MIT