> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cryptechtoday.com/llms.txt
> Use this file to discover all available pages before exploring further.

# LangChain & LLM Agent Toolkit

> Equip Python LangChain, CrewAI, and AutoGen agents with on-chain due diligence tools

# LangChain & LLM Agent Integration

Enable autonomous AI agents to audit smart contracts, check market regimes, and calculate DEX slippage prior to executing financial actions.

***

## Python Tool Declaration (LangChain)

```python theme={null}
import os
import requests
from langchain.tools import tool

CRYPTECH_API_KEY = os.getenv("CRYPTECH_API_KEY")

@tool
def audit_crypto_token(contract_address: str, chain: str = "ethereum") -> str:
    """
    Audits a cryptocurrency token for honeypots, trading taxes, and AMM cashout liquidity.
    Use this tool whenever a user asks to evaluate a token or check if a contract is safe.
    """
    url = "https://api.cryptechtoday.com/api/v1/analyze"
    headers = {
        "X-API-Key": CRYPTECH_API_KEY,
        "Content-Type": "application/json"
    }
    
    res = requests.post(url, json={"address": contract_address, "chain": chain}, headers=headers)
    if res.status_code != 200:
        return f"Audit failed: {res.text}"
    
    d = res.json()
    return (
        f"Token: {d.get('name')} ({d.get('symbol')})\n"
        f"Safety Score: {d.get('total_score')}/100 [{d.get('risk_label')}]\n"
        f"Verdict: {d.get('verdict_summary')}\n"
        f"Max Safe Exit: ${d.get('sellability', {}).get('max_safe_exit_usd', 0):,} USD"
    )

@tool
def get_macro_market_regime() -> str:
    """
    Returns current crypto market macro regime (BULLISH_EXPANSION, BEARISH_CONTRACTION, etc.)
    and Fear & Greed sentiment index.
    """
    res = requests.get("https://api.cryptechtoday.com/api/markets/intel/regime")
    d = res.json()
    return f"Regime: {d.get('regime')} | Stance: {d.get('risk_posture')} | F&G: {d.get('fear_greed_index', {}).get('value')}"
```

***

## Usage in Agent Chain

```python theme={null}
from langchain_openai import ChatOpenAI
from langchain.agents import initialize_agent, AgentType

llm = ChatOpenAI(model="gpt-4o", temperature=0)
tools = [audit_crypto_token, get_macro_market_regime]

agent = initialize_agent(
    tools=tools,
    llm=llm,
    agent=AgentType.STRUCTURED_CHAT_ZERO_SHOT_REACT_DESCRIPTION,
    verbose=True
)

agent.run("Should I invest in 0x514910771AF9Ca656af840dff83E8264EcF986CA on Ethereum right now?")
```


## Related topics

- [AI & Model Context Protocol (MCP)](https://docs.cryptechtoday.com/ai-mcp/claude-cursor-mcp.md)
- [Introduction](https://docs.cryptechtoday.com/index.md)
- [Holder Concentration](https://docs.cryptechtoday.com/api-reference/token-holders.md)
- [Quickstart](https://docs.cryptechtoday.com/quickstart.md)
- [Market Anomaly Alerts](https://docs.cryptechtoday.com/api-reference/market-alerts.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.