# Staking & Security

OpenGPU uses crypto-economic security to ensure honest behavior without centralized enforcement.

## How It Works

Both clients and providers stake $OGPU on-chain. Stakes act as collateral—similar to Proof-of-Stake systems.

The protocol enforces minimum stake requirements, but participants can stake more to signal trustworthiness to counterparties.

### For Providers

| Action                  | Consequence                |
| ----------------------- | -------------------------- |
| Complete tasks honestly | Earn payments + reputation |
| Deliver invalid results | Stake slashed              |
| Fail to deliver         | Stake slashed              |
| Consistent performance  | Better reputation          |

### For Clients

| Action                        | Consequence                  |
| ----------------------------- | ---------------------------- |
| Publish Source                | Stake locked as collateral   |
| Source is malformed/malicious | Stake slashed                |
| Publish Task                  | Payment locked in escrow     |
| Task completed                | Payment released to provider |
| Task expires/cancelled        | Payment refunded             |

## Escrow System

When a client publishes a task:

1. Payment is locked in the Vault contract
2. Providers compete to execute
3. On valid completion, payment releases automatically
4. On failure or timeout, funds return to client

No intermediary holds funds. The protocol enforces the rules.

## Container Security

Providers execute client-provided Docker containers, which introduces trust considerations. The protocol mitigates this through:

* **Client staking**: Clients lock $OGPU when publishing sources. Malicious or malformed containers result in slashing.
* **Sandbox isolation**: Provider Suite runs containers in isolated environments.
* **Provider discretion**: Providers evaluate sources before registering. Higher client stakes and reputation signal trustworthiness.
* **Security agents**: Automated scanning of sources for malicious patterns (see [AI Agents](/opengpu-network/introduction/ai-agents.md)).
* **Enterprise options**: Verified clients and private deployments for higher-trust scenarios.

## Reputation

Provider performance is tracked on-chain:

* Success rate
* Response times
* Task completion history

Higher reputation increases visibility and earning potential.

## Why This Matters

* **No trust required**: Unknown parties can transact safely
* **Self-enforcing**: Bad actors are penalized automatically
* **Transparent**: All activity verifiable on-chain

> See [Token](/opengpu-network/blockchain-reference/token.md) for $OGPU details.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://opengpu-network.gitbook.io/opengpu-network/introduction/staking-and-security.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.