skills.rest
davepoon
davepoonCOMMUNITY
Explore all skills by this author →
Website𝕏@davepoonGitHub

hard-predict-future

Generates quantitative foresight reports on future outcomes using a 12-step Python pipeline.

3.4k|487|Updated Jul 25, 2025
Search Tags:#probability#forecasting#Python#scenario planning#prediction#foresight#steeep analysis
One-click install
npx skills add https://github.com/davepoon/buildwithclaude --skill hard-predict-future
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hard-predict-future
Source: https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/hard-predict-future
Command: npx skills add https://github.com/davepoon/buildwithclaude --skill hard-predict-future

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Answering high-stakes questions about future outcomes ("Will X happen?", "Who will win X?") usually relies on gut feeling or shallow speculation. This Skill replaces guesswork with a deterministic 12-step pipeline that combines web research, signal scoring, historical analogues, and computed probabilities into a structured foresight report.

Core Features & Use Cases

  • Deterministic 12-Step Pipeline: Claude handles intelligence gathering while Python scripts handle all arithmetic — signal scoring, STEEEP matrix building, probability calculation, confidence scoring, and report formatting.
  • Structured Scenario Planning: Produces PROBABLE, PLAUSIBLE, POSSIBLE, and PREFERABLE scenarios with independent 0-100 scores, plus IFTF backcasting and per-stakeholder win conditions.
  • Input Validation & Regional Lens: Rejects invalid queries (fictional entities, geological timescales, private individuals) and applies regional multipliers for India, USA, Europe, and China.
  • Use Case: Ask "Will electric vehicles dominate India's car market by 2035?" and receive a full report with signal pulse, STEEEP matrix, historical analogues, confidence score, and decision guidance.

Quick Start

Ask Claude to run the hard-predict-future skill on a specific future-oriented question such as "Will AI coding assistants replace junior developers by 2030?"

Frequently Asked Questions about hard-predict-future

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict future outcomes with quantitative analysis?▼

Run the hard-predict-future skill with a specific question like "Will X happen by 2030?". It collects web signals, scores them deterministically in Python, finds historical analogues, and outputs probability scores for probable, plausible, and possible scenarios.

What is STEEEP scenario planning and how is it used here?▼

STEEEP classifies evidence into Social, Technological, Economic, Environmental, Ethical, and Political categories across Operational, Strategic, and Civilizational time horizons. The skill builds a 6x3 matrix from scored signals to identify hot zones, blind spots, and cross-impact convergence.

Does hard-predict-future work on claude.ai web?▼

No, it requires the Bash tool and Python 3.x to execute its deterministic scripts, so it only works in Claude Code or environments with shell access. Use the Soft Predict Future variant for claude.ai web instead.

What kinds of prediction questions get rejected?▼

The input validator rejects queries about fictional characters, private individuals, non-existent governance systems, timescales beyond 30 years, and vague questions under five words without a named entity or domain.

How is the confidence score calculated?▼

Confidence combines signal density (0-40 points), evidence balance between supporting and opposing signals (0-30), and historical analogue similarity (0-30), minus a penalty of up to 15 points for blind spots in the STEEEP matrix.

Do the probable, plausible, and possible scores sum to 100?▼

No, each scenario receives an independent 0-100 score based on futures cone methodology. A scenario can score high on multiple types simultaneously since they measure evidence strength, not mutually exclusive probabilities.

Related Skills

Explore More Data & Analytics Skills→
luisschmitzheadlineluisschmitzheadline
23

startup-trend-prediction

Identify and quantify market-trend signals to forecast 1-2 years ahead.

Community
Advanced
GRCEngClubGRCEngClub
367

risk-register-manager

Manage organizational risk registers and generate board-ready risk reports.

Official
Intermediate
kayzaakayzaa
5

Yield Curve Analyzer

Monitor yield curves across 40 global economies and detect inversions.

Community
Advanced
expectedparrotexpectedparrot

create-interview-guide

Generate EDSL Survey interview guides using QuestionInterview and Geiecke & Jaravel methodology.

Official
Intermediate