← 返回 Skills 市场

AI Displacement Monitor

Name: AI Displacement Monitor
Author: spyfree

作者 Li Xin · GitHub ↗ · v1.0.2

cross-platform ✓ 安全检测通过

573

总下载

当前安装

版本数

在 OpenClaw 中安装

/install ai-displacement-monitor

功能描述

Monitor early-warning signals of AI-driven white-collar labor displacement and macro-financial spillovers. Use when you need a practical indicator framework,...

使用说明 (SKILL.md)

AI Displacement Monitor

Use this skill to produce a structured risk monitor for AI-led labor substitution and downstream financial stress.

Output Format

Always return:

Signal Board (10 indicators with latest value, direction, threshold status)
Composite Risk Light (GREEN / YELLOW / ORANGE / RED)
Actionable Notes (portfolio/risk posture suggestions)
Data Gaps (missing or stale inputs)

Indicator Framework

Read references/thresholds.example.json and follow its indicator IDs, thresholds, and tiering.

Also apply the "Industrial-Revolution Lens" when interpreting risk:

Do not evaluate layoffs alone.
Compare substitution speed vs re-absorption speed (new demand + new capex).
If substitution weakens labor but capex/reinvestment accelerates, avoid over-escalating crisis labels.
Tier A (Leading labor demand): A1-A4
Tier B (Labor market confirmation): B1-B3
Tier C (Spillover: consumption/credit): C1-C3

Composite Rule

YELLOW: Tier A triggered >= 2
ORANGE: Tier A >= 2 and Tier B >= 1
RED: Tier A >= 2 and Tier B >= 1 and Tier C >= 1
GREEN: otherwise

Weak-Links Interpretation (Jones Lens)

When assessing macro impact, apply a weak-links check:

Broad automation can still deliver gradual macro gains if key bottleneck tasks remain scarce.
Do not infer immediate macro collapse from partial task automation alone.
If bottleneck proxies remain tight (D3 worsening, D4 weak reinvestment), keep risk elevated.
If bottlenecks ease via reinvestment/capex and purchasing power improves (D1/D2), avoid over-escalation.

Minimum Quality Rules

Time-stamp each metric and note frequency mismatch (weekly vs monthly vs quarterly).
If source coverage is partial, mark confidence as low or medium.
Never hide missing data; list it under Data Gaps.
If more than 3 indicators are missing, downgrade confidence by one level.

Recommended Alert Style

Keep alerts short and decision-oriented:

"What changed"
"Why it matters now"
"What to do next"

Optional JSON Mode

If user asks for machine-readable output, return:

asOf
signals[] (id, value, unit, threshold, triggered, trend)
composite
confidence
gaps[]
notes[]

安全使用建议

This skill is a coherent, instruction-only framework (no installs or secrets). Before using: (1) confirm how your agent will source indicator data—if it will call external APIs or scrape sites, decide which API keys/credentials you will supply and to which component; (2) review and, if needed, adapt the thresholds.example.json to match your data definitions and coverage; (3) if you plan to automate alerts, test outputs manually first (check timestamps, frequency mismatches, and data gaps as the skill recommends); (4) avoid granting broad platform credentials unless necessary—prefer scoped API keys for specific data feeds; and (5) if the agent implementation scrapes login-protected sites (LinkedIn, job boards), be aware of terms-of-service and privacy implications.

功能分析

Type: OpenClaw Skill Name: ai-displacement-monitor Version: 1.0.2 The skill bundle is benign. It defines an AI displacement monitor, providing instructions for an AI agent in `SKILL.md` on how to interpret data, apply rules, and format output. The agent is instructed to read `references/thresholds.example.json`, which is a local data file containing the specific indicator thresholds and rules. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts designed to make the agent perform unauthorized or harmful actions beyond its stated purpose.

能力评估

✓ Purpose & Capability

Name/description match the included indicator definitions and composite rules. The skill is instruction-only and uses the bundled thresholds.example.json for indicator IDs, tiers, triggers and interpretation; it does not request unrelated binaries, credentials, or filesystem access.

ℹ Instruction Scope

SKILL.md is narrowly scoped to producing an indicator board, composite risk light, notes and gaps and explicitly references the bundled thresholds file. It does not instruct the agent to read unrelated system files or environment variables. One ambiguity: the instructions assume the agent will obtain values for the indicators from external data sources (job boards, JOLTS, LinkedIn, market data) but do not specify how to fetch them or which credentials to use—this is a benign omission but means data-acquisition behavior depends on the host agent/integration.

✓ Install Mechanism

No install spec and no code files—this is an instruction-only skill. Nothing is written to disk or fetched during installation.

ℹ Credentials

The skill requests no environment variables or credentials, which is proportional to an indicator/interpretation framework. However, practical use will typically require access to third-party data (APIs or web scraping). The skill itself does not request those credentials, so the user or host agent must supply them separately; verify any credentials you provide are only given to trusted integrations.

✓ Persistence & Privilege

always is false and the skill does not request persistent system-wide configuration or modify other skills. Autonomous invocation is allowed by default but that is standard and not by itself concerning here.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install ai-displacement-monitor
安装完成后，直接呼叫该 Skill 的名称或使用 /ai-displacement-monitor 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.2

Add Weak-Links (Jones) interpretation layer for bottleneck-aware macro risk reading

v1.0.1

Add Industrial-Revolution Lens and 4 balancing indicators for substitution vs re-absorption dynamics

v1.0.0

Initial release: 10-indicator framework, composite risk light rules, and monitoring thresholds

元数据

Slug ai-displacement-monitor

版本 1.0.2

许可证 —

累计安装 1

当前安装数 1

历史版本数 3

常见问题