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

# verification • AI Agent SDK

> Verification Hooks Protocol for PraisonAI Agents.

# verification

<Badge color="blue">AI Agent</Badge>

Verification Hooks Protocol for PraisonAI Agents.

Provides protocols for verification hooks that can be used with Agent autonomy.
Verification hooks run after file writes or at configured checkpoints to
validate agent actions (e.g., run tests, lint, build).

Usage:
from praisonaiagents.hooks import VerificationHook, VerificationResult

class TestRunner(VerificationHook):
name = "pytest"

def run(self, context=None):

# Run tests and return result

return VerificationResult(
success=True,
output="All tests passed",
details=\{"tests\_run": 10, "passed": 10}
)

agent = Agent(
instructions="...",
autonomy=True,
verification\_hooks=\[TestRunner()]
)

## Import

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents.hooks import verification
```

## Classes

<CardGroup cols={2}>
  <Card title="VerificationResult" icon="brackets-curly" href="../classes/VerificationResult">
    Result of a verification hook execution.
  </Card>

  <Card title="VerificationHook" icon="brackets-curly" href="../classes/VerificationHook">
    Protocol for verification hooks.
  </Card>

  <Card title="BaseVerificationHook" icon="brackets-curly" href="../classes/BaseVerificationHook">
    Base class for verification hooks.
  </Card>

  <Card title="CommandVerificationHook" icon="brackets-curly" href="../classes/CommandVerificationHook">
    Verification hook that runs a shell command.
  </Card>
</CardGroup>