ClawHub
Back to skill

Security audit

工具调用重试免费版

Security checks for vulnerabilities and agentic risk

Overview

The skill is not overtly malicious, but it encourages agents to retry broad write-capable actions such as payments, deletes, saves, and exports without strong user-control safeguards.

Review before installing. Use this only for read-only or clearly idempotent tool calls unless you add explicit safeguards such as idempotency keys, deduplication, user confirmation, and retry exclusions for payments, deletes, saves, exports, and other state-changing operations.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill advertises very broad capability and scene keywords without clear trigger boundaries, increasing the chance an agent will invoke it in contexts the user did not explicitly intend. In an agent ecosystem, overly generic matching can route unrelated tasks into a skill that has exec/write access, creating unnecessary exposure to side effects or misuse.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The markdown explicitly supports retry behavior for create/modify/delete, save, and export-style operations but does not present a strong warning that retries can repeat non-idempotent actions and alter data or external state. Although idempotency is mentioned elsewhere, the surrounding guidance normalizes write-capable retries and may lead users or agents to apply them unsafely to payments, updates, deletions, or file writes.

Static analysis

No suspicious patterns detected.