Character Manager

v1.0.4

小说角色管理工具。创建、编辑、查询角色档案,支持角色关系网络、情感弧光、动机追踪。Use when: Architect 代理在 Phase 3 需要创建角色档案、管理角色关系、追踪角色成长轨迹。

0· 157·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name, description, provided templates/configs, example YAMLs, reference guide, and the included Python script all align: this is a local character-management tool for novel projects. There are no unexpected dependencies or credentials requested that don't fit the stated purpose.
Instruction Scope
SKILL.md instructs installing Python deps and running the included script with project paths. It explicitly documents that the tool reads/writes YAML files in the provided --project directory and warns users not to point it at system or sensitive directories. The runtime instructions do not ask the agent to read environment vars or contact external endpoints.
Install Mechanism
There is no automated install spec; instructions recommend running 'pip3 install -r requirements.txt --user' (requirements are pyyaml and rich). This is a standard, low-risk install pattern from PyPI and the package list is proportional to the repository's needs.
Credentials
The skill declares no required environment variables, no primary credential, and no required config paths beyond its own configs. The files and dependencies listed are consistent with local YAML manipulation and CLI output formatting.
Persistence & Privilege
always is false (no forced inclusion). The skill does not request permanent elevated privileges or modify other skills. It can be invoked autonomously (platform default), which is appropriate for this type of tool.
Assessment
This skill appears coherent and self-contained, but be cautious when choosing the --project path: run it against your novel workspace, not system or sensitive directories. If you plan to allow an agent to call this autonomously, review the script's create/update/delete code paths (especially the delete operation) before granting it access, and consider running it in a sandboxed workspace or a git-tracked project so accidental overwrites can be reverted.

Like a lobster shell, security has layers — review code before you run it.

latestvk978xmwvvzxsdwj7zwawt2h9bn8413jv

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments