Back to skill

Security audit

Relentless Goal-Executor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed goal-execution workflow with broad but purpose-aligned tool use and confirmation controls.

Install this if you want a Chinese-language goal-execution assistant that can keep a progress ledger and, if you opt in, resume by heartbeat. Before using it, choose the ledger path deliberately and keep the default strict confirmation mode unless you are comfortable pre-authorizing the listed write or external actions.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (7)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list includes the very broad everyday term “死磕”, which can be invoked in ordinary conversation unrelated to this skill. Because this skill is user-invocable and designed for persistent, tool-using execution, accidental activation could cause unintended planning, file operations, web searches, or workflow persistence to begin in contexts where the user did not intend to launch the skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The natural-language instructions governing the skill are entirely in Chinese, while the skill is marked user-invocable and does not state that it is limited to Chinese-speaking users or offer a language-selection option. That creates a language policy concern because the operational behavior is effectively constrained to a specific language without user opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The file documents capabilities including '文件读写' and '代码执行' and instructs performing minimal validation calls, but it does not include any warning to users that these actions may modify files or run commands in the environment. For markdown files, the rule requires warnings when described behavior could affect user data or system integrity.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This markdown file presents all operational instructions and task descriptions exclusively in Chinese, and nowhere indicates that the user can choose another language or that the skill is intentionally limited to a Chinese-speaking context. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This markdown file presents all headings, instructions, and task labels exclusively in Chinese. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is clearly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The requirement explicitly says the ledger should be automatically read and written after a path is agreed, without per-round confirmation. That creates a persistent file modification behavior that can surprise users, overwrite existing content, or normalize unattended writes across iterations and heartbeats, especially in a skill designed for relentless multi-step execution.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown template mandates a single language for all headings and usage instructions, but does not indicate that Chinese is optional or required for a region-specific purpose. That can violate language/locale policy because users who do not opt into Chinese are forced into one locale.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.