Secure Your First AI Agent in 10 Minutes

Go to clevername.net and click Get Started. Enter your email and a password. Check your inbox for the verification email and click the confirmation link.

In the dashboard, go to Settings → Keys. Click Add Key, select your provider (Anthropic, OpenAI, or Google), and paste your API key. Clevername stores it in GCP Secret Manager — the raw key is never stored in the database.

Go to Agents in the sidebar and click Create Agent. Clevername issues that agent a cn-live-*gateway token automatically — this is what your code will use to authenticate, and it is what ties every request back to a governed identity. If you only need a token for a script rather than a governed agent, Settings → Keys → Personal Access Keys issues cnk_* keys, which the gateway also accepts.

cn-live-a1b2c3d4e5f6g7h8i9j0...

If you use Claude Code, Cursor, VS Code, or Windsurf, install the Guard CLI to get security scanning in your IDE too:

npm install -g clevername
clevername guard setup claude-code --token cn-live-your-token-here
# or: cursor, windsurf, vscode

pip install langchain-openai openai

Change two things in your LangChain code: the base_url and the api_key. No session headers, no registration call — that's it. Your token carries agent identity automatically.

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="claude-sonnet-5",          # or gpt-4o, gemini-pro, etc.
    base_url="https://clevername.net/api/hub/v1",
    api_key="cn-live-your-token-here",  # pragma: allowlist secret
)

# Use it exactly like before
response = llm.invoke("Summarize the key benefits of agent governance.")
print(response.content)

That is it. Your LangChain agent now runs through CleverGuard. Every message is scanned for prompt injections, PII, and secrets before reaching the AI provider.

For CrewAI, set the environment variables and CrewAI routes through Clevername automatically:

import os
from crewai import Agent, Task, Crew

os.environ["OPENAI_API_BASE"] = "https://clevername.net/api/hub/v1"
os.environ["OPENAI_API_KEY"] = "cn-live-your-token-here"  # pragma: allowlist secret

researcher = Agent(
    role="Research Analyst",
    goal="Find and summarize key information",
    backstory="You are an expert researcher.",
    llm="claude-sonnet-5",
)

task = Task(
    description="Summarize the benefits of AI agent governance.",
    expected_output="A 3-bullet summary.",
    agent=researcher,
)

crew = Crew(agents=[researcher], tasks=[task])
result = crew.kickoff()
print(result)

from openai import OpenAI

client = OpenAI(
    base_url="https://clevername.net/api/hub/v1",
    api_key="cn-live-your-token-here",  # pragma: allowlist secret
)

response = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[
        {"role": "user", "content": "What is agent governance?"}
    ],
)
print(response.choices[0].message.content)

Go back to the Clevername dashboard and check:

• Agents → Audit— your call is logged here with model, tokens used, cost, the agent identity that made it, and the scan verdict. This is the verification surface on every plan, including Free.
• Agents → Overview— open the agent you just created to see its health score and the guardrail profile its traffic is being measured against