Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward goal-planning helper with no evidence of hidden execution, persistence, credential access, or data exfiltration.
Install only if a Chinese-first, template-based planning helper is acceptable. Review generated subtasks before acting on spawn_hint suggestions, especially when they involve web research or other agents.
This markdown file is entirely written in Chinese and does not indicate that language selection is optional or that the skill is intended only for a Chinese-speaking or region-specific audience. Under the policy, forcing a specific language without user opt-in or documented justification is a natural-language policy violation.
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
### 核心文件（必需）
- [x] **SKILL.md** - 技能定义文件
- 路径：`skills/goal-decomposer/SKILL.md`
- 包含：name, version, description, metadata
- 状态：已完善（添加tags字段）
The primary description is written in Chinese and presents the skill behavior in that language without any note about letting users choose their preferred language. Under the policy criteria, forcing a specific language or locale without user opt-in is a natural-language policy violation.
The skill description is written to operate in Chinese without documenting locale constraints or user preference handling, which can force language behavior that does not match the user's request. This is primarily a safety and usability issue: users may misinterpret generated tasks or fail to review them accurately, increasing the chance of mistaken execution in multilingual environments.
The trigger description is broad enough to match many ordinary planning requests, which can cause the skill to activate when the user did not explicitly ask for goal decomposition. In an agent system with spawning and routing capabilities, unintended invocation can alter task flow, introduce unnecessary automation, or cause downstream actions to be proposed or initiated from an overly generic prompt.
This JSON example specifies the goal, context, and constraints entirely in Chinese, indicating the skill content is tailored to a single language/locale without any visible opt-in or alternative. Under the policy, forcing a specific language without user choice can be a natural-language policy violation unless the locale limitation is clearly justified.
This JSON contains user-facing goals, constraints, task titles, descriptions, and spawn hints 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 documented and justified, which is not present here.
The manifest describes decomposing high-level natural-language goals into executable multi-level task lists, especially for vague or complex goals. In code, decomposition is driven only by coarse keyword classification and hardcoded templates, and the context parameter is unused, so actual behavior is much narrower than the claimed adaptive goal decomposition.
This JSON skill content is entirely written in Chinese, including the input goal, context, task titles, descriptions, and KPI strings. Because there is no accompanying indication that the skill is region-specific or that the user can choose their preferred language, it may violate a language/locale policy requiring user choice or explicit justification.
This markdown file presents all instructions and examples exclusively in Chinese, which can constitute a language/locale policy issue when no user opt-in or justification is provided. The policy specifically flags skills that force a specific language without offering a choice.
SQP-3 applies to all file types and covers language or locale policy violations. This skill file forces a specific language for the user-facing content and examples, but does not provide any opt-in, alternative language, or justification for restricting the skill to Chinese.
This markdown file presents all instructions and outputs in Chinese and does not indicate that the user can choose another language. Under the policy criteria, forcing a specific language without user opt-in is a natural-language policy concern.
The file uses Chinese-only natural-language descriptions and keyword logic, including the title and docstrings, and classification depends on Chinese trigger words. Under the policy, forcing a specific language without user opt-in can be a locale/language policy violation.
The docstring 拆解目标为任务树 suggests substantive goal-driven decomposition. However, the function only classifies by keywords, selects a preset template, and fills it without using the context argument or adapting subtasks to the actual goal, creating a mismatch between documented intent and behavior.