FB Ads Copywriter Pro
v1.0.1AI 营销文案生成器 - 生成高转化率广告文案、社交媒体内容、邮件营销内容
⭐ 0· 36·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description match the provided code and SKILL.md: the script builds prompt templates and returns example outputs for ad/social/email copy. _meta.json and README mention an OpenAI-compatible client / Qwen integration, but the included script currently only simulates generation (no network calls). This is a minor mismatch but explainable (stubbed implementation).
Instruction Scope
SKILL.md instructs running the included script with arguments; instructions do not ask the agent to read unrelated files, access system config, or send data to external endpoints. The runtime instructions are narrowly scoped to copy generation.
Install Mechanism
No install spec is present (instruction-only). A requirements.txt is included for optional pip install of openai client; there are no downloads or extracts. Risk from installation is low.
Credentials
The skill declares no required environment variables or credentials. The README/_meta mention OpenAI/Qwen usage which would normally require API keys, but the shipped script does not use them. If you later enable real API calls, providing credentials would become necessary—currently absence of env var requests is consistent with the stubbed implementation.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or system settings, and is user-invocable. It has no elevated persistence or privileges.
Assessment
This skill appears to be a local, stubbed copywriter that generates prompt templates and example outputs; it does not contact external APIs or request secrets in its current form. Things to consider before installing: 1) Origin/authenticity — registry metadata shows an owner ID and a GitHub URL in _meta.json but the package registry lists no homepage; verify the source if you need production use. 2) If you plan to enable real model calls (Qwen/OpenAI), you'll need to modify the script and supply API keys — only provide keys after auditing the code and storing keys in secure env vars, not inline. 3) Test in an isolated environment first (no sensitive data) and monitor network traffic to confirm no unexpected exfiltration. 4) If you require remote inference, inspect any code changes that add network calls or third-party dependencies (openai client) for unexpected behavior.Like a lobster shell, security has layers — review code before you run it.
latestvk97e897s2ch3mpgv6pym6ccdvs842bw9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.