antarraksha-crewai 0.1.4

Antarraksha AI Agent Enforcement SDK for Crewai

antarraksha-crewai

Antarraksha AI Agent Enforcement SDK for CrewAI.

Installation

pip install antarraksha-crewai

Quick Start

Instantiate the Antarraksha guardrail and call its check methods from inside your CrewAI tasks or tool wrappers. Registration happens automatically on first use — no login, no API key, no signup required.

from antarraksha_crewai import AntarrakshaGuardrail

guardrail = AntarrakshaGuardrail(agent_id="my-crew-agent")

result = guardrail.check_tool_call(tool_name="search_web", parameters={"query": "..."})
if result["decision"] == "DENY":
    raise PermissionError(result["reason"])

result = guardrail.check_llm_call(model_name="gpt-4o", prompt_length=512)
if result["decision"] == "DENY":
    raise PermissionError(result["reason"])

Wire check_tool_call into your tool callbacks and check_llm_call into your LLM wrapper. Every intercepted call is logged and visible at https://antarraksha.ai/registry.

Parameters

Parameter	Default	Description
agent_id	None	Unique identifier for your agent. Triggers auto-registration on first use.
base_url	"https://antarraksha.ai"	Antarraksha endpoint. Override for self-hosted / dev.
passport_id	None	Optional pre-issued passport ID (e.g. ANTK-PASS-xxx).
sdk_key	None	Optional pre-issued SDK key. If omitted and agent_id is set, the SDK auto-registers and obtains one.
fail_closed	True	If True, deny on enforcement-server unreachable. Set False for fail-open during early integration.

Corporate Network Note

If you're behind a corporate TLS-inspection proxy and see SSLCertVerificationError, install:

pip install pip-system-certs

This makes Python trust your Windows / macOS system certificate store.