Wanxiang Scroll Personality

万象绘卷人格系统分卷 - 包含人格系统、输出规则、质量控制、创作工具

Security Scan

This is an instruction-only persona/formatting pack for a storytelling system; it asks for no credentials or installs and its files are consistent with the declared purpose.

Purpose & Capability

Name/description describe a storytelling/personality system and the included documents provide persona rules, output formats, quality-control and creative tools. There are no unrelated required binaries, env vars, or install steps; the requested surface is proportional to the stated purpose.

Instruction Scope

SKILL.md contains only a short description and install hint for the main package. The reference files contain detailed persona anchors and absolute-sounding rules (e.g. 'CORE_MEMORY_LOCK: ... 不可篡改', strict '绝对禁止' rules and output templates). Those are consistent with a persona pack, but if the agent treats them as immutable runtime constraints they could change assistant behaviour significantly. The documents explicitly say they are "创作参考材料" and not runtime instructions, which reduces but does not eliminate the risk of overreach if the agent applies them as policies.

Install Mechanism

No install spec, no code files, and nothing is downloaded or written to disk. This is the lowest-risk install model (instruction-only).

Credentials

No environment variables, credentials, or config paths are requested. Nothing in the files asks for secrets or unrelated system access.

Persistence & Privilege

The skill does not request permanent presence (always:false) nor modify agent configs. However, the content encourages 'memory anchoring' and '不可篡改' core memories in its prose; that is a behavioral directive embedded in docs rather than a platform-level persistence flag. If an agent applies these anchors persistently, that changes agent behaviour but the skill itself does not request platform privileges.

Assessment

This is a documentation-only persona pack for fiction/storytelling — it does not request credentials or install anything. Before installing, consider: (1) The documents include strong persona anchors and absolute-sounding rules that could change assistant behaviour (e.g., 'CORE_MEMORY_LOCK' and '绝对禁止' clauses). If you run agents that must follow strict safety/moderation, test the skill in a sandbox first. (2) The material contains sexual/explicit design elements (e.g., '性能力', '高潮判定') and explicit style rules; confirm this matches your content policies. (3) Because it can alter output style and persona, avoid granting it access to sensitive systems or credentials; if you're concerned about an agent adopting persistent behaviours, disable autonomous invocation or thoroughly review how your agent applies instruction-only reference files. If you need more certainty, ask the skill author for a safety summary and explicit statement about whether these files are meant to be treated as immutable runtime constraints.

万象绘卷 - 人格系统分卷

人格架构 + 输出规则 + 创作风格 | v1.0.0

包含内容

第十章 - 人格系统与输出规则

• 人格系统架构
• 输出规则定义
• 质量控制标准
• 创作工具集

使用方式

配合主技能包 wanxiang-scroll 使用：

clawhub install wanxiang-scroll
clawhub install wanxiang-scroll-personality

---

万象绘卷分卷系列 - 人格系统