ClawHub
Back to skill

Security audit

Implement Issue

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Gitee issue-to-PR workflow helper, with expected remote project actions but no evidence of hidden or unrelated behavior.

Install this only if you want an agent to work against Gitee issues and perform visible repository actions such as posting issue comments, creating pull requests, and updating issue status. Review the issue and repository target before allowing write actions.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill metadata presents this as an implementation helper, but the documented workflow also performs external state-changing actions such as commenting on issues, updating issue status, and creating pull requests. That scope expansion can cause users or higher-level agents to invoke the skill expecting local coding assistance while unintentionally authorizing remote project-management changes.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrases are broad enough to match ordinary development requests like 'work on this issue' or 'implement issue', increasing the chance that the skill is auto-selected in contexts where the user did not intend remote Gitee operations. Because this skill can create PRs and modify issue state, overbroad activation materially raises the risk of unintended external actions.

Intent-Code Divergence

Low
Confidence
82% confidence
Finding
The prerequisites state the required Gitee tools are `get_repo_issue_detail`, `list_repo_issues`, `create_pull`, `comment_issue`, and `update_issue`, but Step 1 later instructs use of `list_issue_comments`. This is a direct contradiction between the declared required toolset and the documented procedure.

Static analysis

No suspicious patterns detected.