Strategy Optimizer Agent
v1.0.0策略优化智能体。当需要优化创作策略、调整发布节奏、分析ROI、沉淀方法论时触发本技能。用于：(1) 评估当前策略有效性 (2) 提出选题/发布/文风优化建议 (3) 计算投入产出比 (4) 沉淀方法论到知识库 (5) 预测市场趋势
⭐ 0· 34·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description match the SKILL.md content: the file describes assessing strategy, recommending topic/publishing/style changes, ROI calculation, knowledge-base outputs, and forecasting. There are no unrelated requirements (no env vars, binaries, or installs) that would be disproportionate to the stated purpose.
Instruction Scope
Runtime instructions are high-level, data-driven guidelines and output templates (JSON/markdown). They assume receiving reports from human roles (market/datascience) rather than instructing the agent to read arbitrary system files or credentials. The instructions are structured and do not ask the agent to exfiltrate data or call external endpoints.
Install Mechanism
No install spec and no code files are present — instruction-only skill with no artifacts written to disk. This is the lowest-risk install profile.
Credentials
The skill declares no required environment variables, credentials, or config paths. All recommended inputs are domain data (reports, historical analyses) provided by humans; there is no disproportionate credential or secret access.
Persistence & Privilege
always:false and default autonomous invocation settings are present. Nothing in the skill requests elevated or permanent platform-wide privileges, nor does it modify other skills or system config.
Assessment
This skill is internally coherent and appears safe in that it only contains guidance/templates and requests no credentials or installs. Before enabling it: (1) verify the source and provenance (metadata shows unknown homepage/author), (2) supply only non-sensitive, de-identified input data (no secrets or raw logs), (3) treat its recommendations as advisory — require human approval for major changes, and (4) test outputs in a controlled environment to confirm the recommendations match your business context and metrics.Like a lobster shell, security has layers — review code before you run it.
latestvk972fr41mpd5pkb92vah1fthfs843pgj
License
MIT-0
Free to use, modify, and redistribute. No attribution required.