marketing-calendar

v1.0.0

华与华营销日历创意生成器。 输入品牌名与品牌资料,自动执行「盘点品牌资产 → 匹配文化母体 → 五要素创意 → 打分筛选」流程, 输出≥80分的营销日历活动方案,并生成可视化 HTML 报告。

0· 628·2 current·2 all-time
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 and description (marketing calendar / 活动创意生成) align with the SKILL.md and marketing-calendar.md contents. The skill requests no binaries, no environment variables, and no installs — which is proportionate for a text-based creative generator that produces an HTML report. Using WebSearch to supplement missing brand data is coherent with the stated behavior.
Instruction Scope
The runtime instructions are narrowly focused on creating marketing activities and an HTML report. Two behavioral choices deserve attention: (1) the skill mandates '不询问、不确认、直接执行' (do not ask, execute directly) and (2) it instructs the agent to '先用 WebSearch 搜索补全' if brand data is insufficient. These are coherent with an automated generator but mean the agent may perform network searches on brand-related terms and will not prompt the user for clarifying info — this could lead to unexpected external queries or outputs if input data is incomplete.
Install Mechanism
There is no install specification and no code files — the skill is instruction-only. That is the lowest-risk install model: nothing is downloaded or written during installation beyond the agent interpreting the SKILL.md.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. Nothing in the instructions asks for secrets or unrelated credentials. This is proportionate for a marketing idea generator.
Persistence & Privilege
always:false (not force-included). The skill will write a local file marketing_calendar_report.html as part of normal operation — this is expected but is a persistent artifact the user should review. The skill's trigger rules mean it may be auto-invoked when users mention certain keywords; autonomous invocation is a platform default, so this is not itself a flaw but worth being aware of.
Assessment
This skill appears internally consistent and does not request credentials or install code, but review these points before installing: (1) The agent is instructed to run without asking follow-up questions and to perform WebSearch when brand data is insufficient — expect network lookups for public brand info and no interactive clarifications. (2) The skill writes a local HTML file (marketing_calendar_report.html); inspect that file before sharing. (3) Do not pass secrets or sensitive internal documents as 'brand_profile' since the skill may include searchable terms in WebSearch. (4) Because the skill's source and homepage are unknown, test it first with non-sensitive or dummy brand data to confirm output quality and behavior. If you want the agent to prompt for missing details, prefer manual invocation or modify the instructions to require clarification steps.

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

latestvk97f2mzp2527tp22hs9re172n581kvp0

License

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

Comments