ClawHub
Back to skill

Security audit

errand-ai

Security checks for vulnerabilities and agentic risk

Overview

This skill clearly dispatches paid local errand workers through the Errand API and requires user confirmation before spending money.

Install only if you are comfortable giving this skill an Errand API key and sending location/task details to errand.be and a human worker. Review the price before approving any dispatch, because confirmed dispatches escrow prepaid KRW and cannot be cancelled after a worker accepts.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

External Transmission

Medium
Category
Data Exfiltration
Content
With a key, confirm the account can afford it:

```bash
curl -s https://errand.be/api/v1/capabilities -H "Authorization: Bearer $ERRANDAI_API_KEY"
# → balance_krw, caps.max_reward_per_task_krw, caps.daily_budget_remaining_krw, task_types[]
```
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The markdown explicitly instructs that `title`, `instructions`, `validation_criteria`, and report labels must be written in Korean. This is a natural-language locale constraint presented as mandatory behavior, and the file does not offer a user choice or explain a justified organizational policy exception.

Static analysis

No suspicious patterns detected.