ClawHub

fiction-crafter

v1.1.0

根据用户提供的内容方向自动生成提示词并创作爽文小说。适用场景：(1) 用户提供小说方向/题材/关键词，(2) 需要生成章节连贯的长篇爽文，(3) 需要维护角色、地点、情节的连续性，(4) 需要为关键情节生成图解，(5) 需要记录生成失败场景以优化后续创作。支持都市、修仙、玄幻、重生、系统流等多种题材。Use wh...

⭐ 0· 126·1 current·1 all-time
by@followaf
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (novel/fiction generator) match the actual contents: templates, .learnings memory files, chapter generation workflow and two small helper scripts. No extraneous environment variables, binaries, or unrelated dependencies are requested.
✓
Instruction Scope
SKILL.md instructs the agent to read/write files inside the skill workspace (.learnings/, output/, assets/) and to call the included scripts. Those file I/O actions are necessary for maintaining story state and formatting chapters; there are no instructions to read system config, shell history, or other unrelated paths, nor to send data to external endpoints.
✓
Install Mechanism
No install spec or remote downloads are present; the skill is instruction-and-script-only. The included scripts are small, local utilities (bash and Python) and there is no evidence of fetching or executing external code.
✓
Credentials
The skill declares no required environment variables, no credentials, and no config paths outside its own workspace. All required files live under the skill directory (.learnings/, output/, assets/), which is proportionate to its stated purpose.
ℹ
Persistence & Privilege
always:false (normal). The skill requires file read/write permission for its workspace and the init script can overwrite/reset .learnings/ and remove output/*.md when run with --clean. This is expected for an authoring tool but worth noting because data in that directory may be changed or deleted.
Assessment
This skill appears to do exactly what it says: generate chaptered fiction while maintaining local 'memory' files. Before installing/using: (1) run it in a dedicated project directory so its .learnings/ and output/ files don't mix with other work; (2) be aware the init script (--clean) overwrites .learnings templates and deletes output/*.md — back up any files you care about before running cleanup; (3) review generated files if you store any sensitive data in the skill folder (the skill will read/write files in that folder); (4) no network credentials are required, and there are no remote installs, but always inspect third‑party skills you add to confirm they match your expectations.

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

latestvk976kwbwsfsegqwq6dkcg36m6984ersr

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments