nerq-langchain

Trust-gate your LangChain agents. Preflight trust checks before every tool call.

Installation

pip install nerq-langchain

Dependencies: langchain-core, requests. Python 3.9+.

trust_gate

Wrap any LangChain agent with automatic preflight trust checks on every tool call.

from nerq_langchain import trust_gate

agent = initialize_agent(tools, llm)
agent = trust_gate(agent, min_trust=60)
# Every tool call now gets a trust check

What happens on each tool call:

from nerq_langchain import trust_gate, TrustError

agent = trust_gate(agent, min_trust=70)
try:
    result = agent.run("Analyze this code")
except TrustError as e:
    print(f"Blocked: {e.tool_name} (trust={e.trust_score})")
    # e.tool_name, e.trust_score, e.recommendation

NerqPreflight tool

LangChain tool that agents can use directly to check trust on any agent or tool.

from nerq_langchain import NerqPreflight

tool = NerqPreflight()
print(tool.run("SWE-agent"))
# Nerq Preflight: SWE-agent
# Recommendation: PROCEED
# Trust Score: 92.5/100
# Grade: A+
# Verified: Yes
# Category: security
# This agent is trusted. Safe to interact.

NerqSearch tool

Find the best agent for any task from 204K+ indexed agents.

from nerq_langchain import NerqSearch

tool = NerqSearch()
print(tool.run("code review"))
# Nerq Search: 'code review' — 5 results
# 1. SWE-agent/SWE-agent — Trust: 92.5, Category: security
# 2. skill-code-review — Trust: 86.0, Category: coding
# ...

Full example

from langchain_openai import ChatOpenAI
from langchain.agents import initialize_agent, AgentType
from nerq_langchain import NerqPreflight, NerqSearch, trust_gate

llm = ChatOpenAI(model="gpt-4")
tools = [NerqPreflight(), NerqSearch()]
agent = initialize_agent(tools, llm, agent=AgentType.OPENAI_FUNCTIONS)

# Trust-gate: block tools with trust < 60
agent = trust_gate(agent, min_trust=60)

result = agent.run("Is SWE-agent safe to use?")

With LangGraph

from nerq_langchain import NerqPreflight, NerqSearch
from langgraph.prebuilt import create_react_agent

tools = [NerqPreflight(), NerqSearch()]
agent = create_react_agent(model, tools)

Why trust-gate your agents?

As AI agents delegate tasks to other agents, every interaction becomes a trust decision. An agent accepting untrusted input or delegating to a compromised tool creates a chain of liability. trust_gate adds a zero-config trust layer: before any tool call executes, Nerq checks the tool's trust score across 204K indexed agents and makes a PROCEED/CAUTION/DENY decision in <50ms. No API key required. Graceful fallback if Nerq is unreachable.

PyPI · Preflight API · KYA — Know Your Agent · Full API Docs