ClawHub

Ecommerce Platform Policy Safe Rewrite

v1.0.0

Rewrite ecommerce ads, creator scripts, landing page lines, or product claims into safer platform-friendly language without killing the selling angle. Use wh...

⭐ 0· 109·0 current·0 all-time
byLeroyCreates@leooooooow
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
The name and description (rewriting risky e-commerce copy into platform-safe variants) match the SKILL.md workflow and output template. No unexpected capabilities or external service integrations are requested.
✓
Instruction Scope
SKILL.md describes extracting claims, highlighting risky lines, and producing safer rewrites and variants. It does not ask the agent to read system files, environment variables, or send data to any external endpoint. The guidance explicitly avoids legal/regulatory advice, which is appropriate.
✓
Install Mechanism
No install spec or code files are included (instruction-only skill). Nothing will be written to disk or fetched at install time.
✓
Credentials
The skill requires no environment variables, credentials, or config paths. Requested inputs are only copy, channel, product category, evidence level, tone, etc., which are proportional to the task.
✓
Persistence & Privilege
always:false and no special privileges requested. disable-model-invocation is false (agent can invoke autonomously) which is the platform default and acceptable here given the limited scope and lack of sensitive permissions.
Assessment
This skill is coherent and contains no hidden installs or credential requests, but keep these points in mind before using it: (1) Do not paste sensitive personal data or confidential business secrets into the copy you submit. (2) For regulated claims (health, finance, medical, legal), have a human expert or legal reviewer validate rewrites — the skill explicitly is not a substitute for formal legal/regulatory review. (3) Provide clear evidence level when asking for stronger vs safer variants, and retain a human-in-the-loop for final ad/creative approval. (4) If you integrate this into an agent workflow, ensure the agent's policies require user confirmation before sending external or sensitive data anywhere.

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

latestvk9763p500jrp8m2ej7aw60ncsh835781

License

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

Comments