ClawHub
Back to skill

Security audit

testing-with-sandbox

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Mailtrap Email Sandbox reference guide with no executable code, persistence, hidden commands, or purpose-mismatched behavior.

Install only if you use Mailtrap Email Sandbox. Keep sandbox tokens separate from production tokens, avoid placing real customer data in test messages unless your team permits it, and confirm any current SDK options against the linked official docs before generating code.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Whitespace Padding

Medium
Category
Prompt Injection
Content
Use [API docs](https://docs.mailtrap.io/developers/) for details, but typical endpoints include:

| Operation       | URL                                                                                          | Reference                                                                                 |
| --------------- | -------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
| List sandboxes  | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes`                          | [Sandboxes API](https://docs.mailtrap.io/developers/email-sandbox/sandboxes-inboxes.md)   |
| List messages   | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes/{inbox_id}/messages`      | [Messages](https://docs.mailtrap.io/developers/email-sandbox/messages.md)                 |
Confidence
70% confidence
Finding
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Whitespace Padding

Medium
Category
Prompt Injection
Content
Use [API docs](https://docs.mailtrap.io/developers/) for details, but typical endpoints include:

| Operation       | URL                                                                                          | Reference                                                                                 |
| --------------- | -------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
| List sandboxes  | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes`                          | [Sandboxes API](https://docs.mailtrap.io/developers/email-sandbox/sandboxes-inboxes.md)   |
| List messages   | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes/{inbox_id}/messages`      | [Messages](https://docs.mailtrap.io/developers/email-sandbox/messages.md)                 |
Confidence
70% confidence
Finding
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Whitespace Padding

Medium
Category
Prompt Injection
Content
Use [API docs](https://docs.mailtrap.io/developers/) for details, but typical endpoints include:

| Operation       | URL                                                                                          | Reference                                                                                 |
| --------------- | -------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
| List sandboxes  | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes`                          | [Sandboxes API](https://docs.mailtrap.io/developers/email-sandbox/sandboxes-inboxes.md)   |
| List messages   | `GET https://mailtrap.io/api/accounts/$MAILTRAP_ACCOUNT_ID/inboxes/{inbox_id}/messages`      | [Messages](https://docs.mailtrap.io/developers/email-sandbox/messages.md)                 |
Confidence
70% confidence
Finding
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Static analysis

No suspicious patterns detected.