# Missing Authorization (CWE-862)

The product does not perform an authorization check when an actor attempts to access a resource or perform an action.

- Prevalence: High Frequently exploited
- Impact: High 3 high-severity rules
- Prevention: Documented 3 fix examples

**OWASP:** Broken Access Control (A01:2021-Broken Access Control) - #1

## Description

Assuming a user with a given identity, authorization is the process of determining whether that user can access a given resource. When authorization checks are completely missing, any authenticated user can access protected resources.

## Prevention

Prevention strategies for Missing Authorization based on 3 Shoulder detection rules.

### Go

Require human approval for sensitive operations and limit agent loop iterations

## Warning Signs

- [HIGH] Excessive LLM agency detected: ...
- [HIGH] LLM implementations with excessive autonomy allowing destructive operations without human approval

## Consequences

- Read Application Data
- Modify Application Data
- Gain Privileges

## Mitigations

- Implement authorization checks for all protected resources
- Use a centralized authorization mechanism
- Follow the principle of deny by default

## Detection

- Total rules: 3
- Languages: go, javascript, typescript, python

## Rules by Language

### Go (1 rules)

- **LLM Excessive Agency** [HIGH]: Detects LLM implementations with excessive autonomy allowing destructive operations without human approval.
  - Remediation: Require human approval for sensitive operations and add iteration limits.

```go
if sensitiveOperations[toolName] {
    queueForApproval(toolCall)
}
```

Learn more: https://shoulder.dev/learn/go/cwe-862/llm-excessive-agency

### Javascript (1 rules)

- **LLM Excessive Agency** [HIGH]: Detects LLM implementations with excessive autonomy that can lead to unintended
consequences. OWASP LLM08 - Excessive Agency.

Excessive agency occurs when LLMs are granted:
- Ability to perform destructive operations without confirmation
- Auto-execution of LLM-generated code or commands
- Direct database modifications without approval
- Financial transactions without human oversight
- Email/communication sending without review

This rule detects:
- Auto-execution of tool calls without human approval
- Destructive operations (delete, update, send) without confirmation
- Autonomous agent loops without termination controls
- Missing human-in-the-loop for sensitive operations
  - Remediation: Require human approval for sensitive operations and set iteration limits on agent loops.

```javascript
if (SENSITIVE_OPERATIONS.has(toolName)) {
  await queueForApproval(toolCall);
} else {
  await executeSafeToolCall(toolCall);
}
```

Learn more: https://shoulder.dev/learn/javascript/cwe-862/llm-excessive-agency

### Typescript (1 rules)

- **LLM Excessive Agency** [HIGH]: Detects LLM implementations with excessive autonomy that can lead to unintended
consequences. OWASP LLM08 - Excessive Agency.

Excessive agency occurs when LLMs are granted:
- Ability to perform destructive operations without confirmation
- Auto-execution of LLM-generated code or commands
- Direct database modifications without approval
- Financial transactions without human oversight
- Email/communication sending without review

This rule detects:
- Auto-execution of tool calls without human approval
- Destructive operations (delete, update, send) without confirmation
- Autonomous agent loops without termination controls
- Missing human-in-the-loop for sensitive operations
  - Remediation: Require human approval for sensitive operations and set iteration limits on agent loops.

```javascript
if (SENSITIVE_OPERATIONS.has(toolName)) {
  await queueForApproval(toolCall);
} else {
  await executeSafeToolCall(toolCall);
}
```

Learn more: https://shoulder.dev/learn/javascript/cwe-862/llm-excessive-agency

### Python (1 rules)

- **LLM Excessive Agency** [HIGH]: Detects LLM implementations with excessive autonomy that can lead to unintended
consequences. OWASP LLM08 - Excessive Agency.

Excessive agency occurs when LLMs are granted:
- Ability to perform destructive operations without confirmation
- Auto-execution of LLM-generated code or commands
- Direct database modifications without approval
- Financial transactions without human oversight
  - Remediation: Require human approval for sensitive operations and set iteration limits.

```python
SENSITIVE_OPERATIONS = {'delete_record', 'send_email', 'process_payment'}
MAX_ITERATIONS = 10

for tool_call in response.choices[0].message.tool_calls:
    if tool_call.function.name in SENSITIVE_OPERATIONS:
        await queue_for_approval(tool_call)
    else:
        await execute_safe_tool_call(tool_call)
```

Learn more: https://shoulder.dev/learn/python/cwe-862/llm-excessive-agency