OAISR - 职业AI替代风险评估

v1.1.2

OAISR - Occupational AI Displacement Risk | 职业AI替代风险评估 中文说明: 职业AI暴露度分析工作流。当用户发送职业名称、询问"XX职业的暴露度"、或提及"AI替代风险"时激活。 输出标准四步分析报告: 1. 双进度条(理论暴露度 vs 实际暴露度) 2. 任务分解表...

0· 100·0 current·0 all-time
bytbook@silent404
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, description, and the runtime instructions are consistent: the skill describes an assessment workflow (dual progress bars, task breakdown, weighted estimate) and does not request capabilities (cloud keys, binaries, dataset mounts) unrelated to that purpose.
Instruction Scope
SKILL.md contains only the analysis workflow, output formatting, data sources, and estimation rules. It does not instruct the agent to read arbitrary system files, access environment variables, or transmit data to unexpected endpoints. It references an external report (Anthropic) as a data source; the instructions expect the agent to cite or use that source but do not include steps that would exfiltrate data or require unrelated system access.
Install Mechanism
There is no install spec and no code files — the skill is instruction-only, so nothing is written to disk and no third-party packages are fetched.
Credentials
The skill requests no environment variables, credentials, or config paths. This is proportionate to an analysis/reporting workflow that is intended to run on agent reasoning without external secrets.
Persistence & Privilege
always is false and the skill does not request elevated or persistent presence. Autonomous invocation is allowed (platform default) but not combined with sweeping privileges or credential access.
Assessment
This skill appears coherent and low-risk because it's instruction-only and asks for no credentials or installs. Before installing, consider: 1) The outputs are estimates based on a cited report — verify citations and, if you need precise numbers, fetch the original Anthropic material yourself. 2) If you provide sensitive, personally identifiable information in job descriptions (names, company-specific context), treat results as potentially logged by the agent—avoid sharing private data unless you trust the agent/runtime. 3) Confirm your deployment environment's network policy: the skill references an external report URL, so the agent might attempt to fetch it if allowed. 4) The skill author is anonymous; if provenance matters, prefer skills from known maintainers or verify the referenced source directly.

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

latestvk973qf76k9nphf3ns98f5qenxh8417c2

License

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

Comments