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

# Security

> Permission, wallet, and execution security model of Blooop.

# Security

Security within Blooop is designed around a single principle:

Autonomy must operate within human-defined intent.

As agents gain the ability to transact, allocate capital, and coordinate actions, security becomes foundational to safe execution.

***

## Security Philosophy

Blooop does not treat agents as trusted actors by default.

Instead, it enforces:

* Explicit permissions
* Controlled execution
* Financial safeguards
* Human approval layers

Agents operate with capability — but never without constraint.

***

## Wallet Ownership Model

Agents require wallets to perform economic actions.

Blooop implements a user-owned custody model.

### Key principles

* Wallets are user-controlled
* Agents are permissioned operators
* Funds remain under user authority
* Execution requires defined approvals

Agents do not “own” funds — they operate within delegated authority.

***

## Permission Framework

Permissions define what agents are allowed to do.

### Configurable controls

* Spending limits
* Asset exposure caps
* Transaction types
* Contract interaction scopes
* Skill authorization

Permissions ensure agents cannot exceed predefined boundaries.

***

## Approval Layers

Sensitive actions require explicit human authorization.

Approval prompts are triggered for:

* High-value transactions
* Treasury allocation
* Smart contract execution
* Cross-chain transfers

Approvals can be executed via dashboard or mobile app.

***

## Execution Constraints

Agents operate within policy-defined constraints.

Constraints include:

* Daily spending caps
* Transaction frequency limits
* Execution timing windows
* Risk exposure thresholds

Execution halts when constraints are exceeded.

***

## Human Intent Enforcement

Blooop embeds human intent directly into execution logic.

Intent enforcement mechanisms:

* Approval gating
* Wallet signature requests
* Spend confirmations
* Permission escalation prompts

Autonomy cannot bypass human-defined rules.

***

## Emergency Controls

Users can intervene instantly if needed.

Emergency capabilities include:

* Pause agents
* Terminate execution
* Lock wallets
* Revoke permissions

These controls protect against unintended behavior.

***

## Skill Security Isolation

Skills operate within restricted environments.

Security isolation ensures:

* Limited wallet access
* Scoped execution rights
* Controlled data permissions

No skill can access capabilities outside its authorization scope.

***

## Auditability & Transparency

All agent activity is logged and observable.

Users can review:

* Transaction history
* Execution logs
* Skill actions
* Approval records

Transparency ensures accountability across autonomous workflows.

***

## Security Stack Summary

Blooop security combines:

* User-owned wallets
* Permission frameworks
* Approval layers
* Execution constraints
* Emergency overrides
* Activity auditing

This creates a controlled environment where autonomous agents can operate safely.

***

## Closing Principle

Blooop does not remove humans from execution — it embeds them at the intent layer, ensuring autonomy remains aligned with user authority.