Copywriter
v1.0.0Professional copywriting for marketing, exec comms, and customer content that uses systematic frameworks and storytelling to deliver clear, audience-focused,...
⭐ 0· 59·0 current·0 all-time
byJon High@jhigh1594
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (elite copywriter for marketing/executive comms) match the content and required actions. The skill's primary behaviors—selecting copy frameworks, consulting templates and storytelling references, and gathering project/brand context—are appropriate for its stated purpose.
Instruction Scope
Runtime instructions explicitly tell the agent to read project documentation (e.g., CLAUDE.md or .claude/CLAUDE.md), search project files for messaging patterns, and request missing audience/product info from the user. This is reasonable for copywriting, but it means the agent may access arbitrary repository files during context gathering; users should be aware the skill will look through project content and may surface sensitive project data if present.
Install Mechanism
No install spec and no code files — instruction-only skill. Low surface area: nothing is downloaded, written to disk, or executed by the installer.
Credentials
The skill requests no environment variables, credentials, or config paths. All declared resource needs are proportional to copywriting (templates and reference docs included).
Persistence & Privilege
always:false and no special OS/config access. The skill does not request persistent presence or modification of other skills or system settings. Normal autonomous invocation is allowed by default but not escalated.
Assessment
This skill appears coherent and low-risk for its stated purpose. The main privacy consideration is that it asks the agent to read project documentation and search project files for context — if your repository contains secrets or sensitive data, the agent could read and incorporate them into outputs. Before installing or enabling autonomous runs, consider: (1) limiting the agent's access to only the specific docs you want it to read, (2) disabling autonomous invocation if you do not want it to run without review, and (3) reviewing any copy outputs for accidental disclosure of confidential information. If you want higher assurance, open the reference files and confirm they don't include secrets, or run the skill only in isolated/test projects.Like a lobster shell, security has layers — review code before you run it.
latestvk973558ztgy89706vyjz9wg6ch83wshc
License
MIT-0
Free to use, modify, and redistribute. No attribution required.