Back to skill

Security audit

Agent Task

Security checks for vulnerabilities and agentic risk

Overview

This markdown-only skill documents a cloud task-management API for agents, with expected authentication, task mutation, attachment, and deletion capabilities disclosed for its purpose.

Before installing, verify that you trust Beijing Guangxian Technology Co., Ltd. with task content, comments, attachments, email identifiers, and workflow history. Treat delete operations as potentially permanent unless the service documentation says otherwise, and require explicit user confirmation before deleting or forwarding important tasks or attachments.

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

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 26)May include surrounding context.

md
**Important**: The interface may change at any time, please obtain the latest definition through the API specification address.

### Authentication Method
Use Authorization header to pass Bearer token for authentication. Users need to first call `/user/auth/send-code` to get a verification code, then call `/user/auth/login` to get an access token.

### Service Address Change
If API retries fail multiple times, please visit [ClawHub](https://clawhub.ai/) to update the skill.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

md
**Important**: The interface may change at any time, please obtain the latest definition through the API specification address.

### Authentication Method
Use Authorization header to pass Bearer token for authentication. Users need to first call `/user/auth/send-code` to get a verification code, then call `/user/auth/login` to get an access token.

### Service Address Change
If API retries fail multiple times, please visit [ClawHub](https://clawhub.ai/) to update the skill.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill explicitly exposes destructive capabilities such as deleting tasks, comments, attachments, and associated data, but it does not warn the user or calling agent about permanence, recovery limitations, or the need for confirmation before execution. In an autonomous multi-agent context, omission of such safeguards increases the risk of unintended data loss through mis-execution, prompt ambiguity, or agent chaining.

Content

No source excerpt is available for this finding.