{
  "markdown": "# Foresight Intelligence\n\n> Strategic foresight engine using IFTF methodology. Two modes: **Soft Predict Future** (Claude-native skill, instant, works on claude.ai) and **Hard Predict Future** (deterministic 12-step pipeline, Python-computed, auditable — requires Claude Code). Structural drivers, cross-impact analysis, IFTF backcasting, four independent futures with per-stakeholder conditional analysis. **Year is optional** — ask any future question and the engine infers the right time horizon.\n>\n> **Author:** Santhosh Gandhi · **Version:** 2.2.0\n\n---\n\n## Try Asking\n\nYear is optional. The engine infers the right horizon from your question.\n\n```\n■ Who will win — Google or Perplexity?\n■ Will OpenAI or Anthropic dominate the AI race?\n■ Will India become the global AI leader?\n■ Will crypto replace banks?\n■ Will remote work become permanent?\n■ Will EVs dominate Indian cities by 2032?\n■ Will UPI become Southeast Asia's default payment rail by 2028?\n■ Will Europe lead the green energy transition by 2035?\n```\n\n---\n\n## Two Modes\n\n| | Soft Predict Future | Hard Predict Future |\n|---|---|---|\n| **How** | Claude-native skill — just ask a question | Say \"run hard predict: [question]\" |\n| **Platform** | Claude Code, claude.ai, Claude for Work | Claude Code only (needs Python + Bash) |\n| **Scoring** | Claude estimates using the formula | Python computes deterministically |\n| **Reproducibility** | ±2–5% variance per run | Identical every run |\n| **Audit trail** | Claude reasoning (implicit) | JSON files for every step |\n| **Best for** | Exploration, quick reads, content | VC memos, high-stakes decisions |\n\n---\n\n## Install on Claude Code\n\n```bash\n# Step 1 — Add the marketplace (one-time setup)\nclaude plugin marketplace add isanthoshgandhi/foresight-intelligence\n\n# Step 2 — Install the plugin\nclaude plugin install foresight-intelligence\n```\n\nThen just ask any future question — Soft Predict activates automatically.\n\nFor Hard Predict Future say:\n```\nRun hard predict: Will India become the global AI leader by 2050?\n```\n\nTo invoke explicitly by name:\n```\n/foresight-intelligence:hard-predict-future Will OpenAI or Anthropic win by 2030?\n```\n\n---\n\n## What You Get\n\nEvery run — both modes — always outputs the same complete report:\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n[SOFT / HARD] PREDICT FUTURE · FORESIGHT INTELLIGENCE\n[Query]\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nPREDICTIONS\n■ Probable  [X/100] [████████████░░░░░░░░] — most likely trajectory\n■ Plausible [X/100] [████████░░░░░░░░░░░░] — credible alternative\n■ Possible  [X/100] [████░░░░░░░░░░░░░░░░] — low-probability but real\n■ Preferable         [stakeholder analysis below]\n\nConfidence: [X]/100 | Signals: [N] | Horizon: [YYYY–YYYY] | [Date]\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nSIGNAL PULSE         — supporting / opposing / wildcard counts + visual bars\nSTEEEP MATRIX        — 6×3 grid with ★ hot  ● warm  ✗ blind indicators\nSTRUCTURAL DRIVERS   — D1, D2, D3 with stability rating\nCROSS-IMPACT         — convergence and friction across time horizons\nHISTORICAL MATCH     — best analogue + similarity score\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n■ PROBABLE  scenario  — narrative + PROOF + IF + BUT + DRIVER\n■ PLAUSIBLE scenario  — narrative + PROOF + IF + BUT + DRIVER\n■ POSSIBLE  scenario  — narrative + PROOF + IF + BUT + DRIVER\n■ PREFERABLE          — IFTF backcasting from desired future to today\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nPREFERABLE FUTURES · Per stakeholder\n  [Player A]: Wins IF → [condition]  BUT ONLY → [constraint]  ONLY THEN → [outcome]\n  [Player B]: Wins IF → [condition]  BUT ONLY → [constraint]  ONLY THEN → [outcome]\n  Users:      Wins IF → [condition]  BUT ONLY → [constraint]  ONLY THEN → [outcome]\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nTHE ONE THING        — the single variable that determines which scenario activates\nDECISION GUIDANCE    — recommended stance, low-regret move, risk trigger\nREGIONAL LENS        — India / USA / Europe / China multipliers\nMETHODOLOGY KEY      — one-line explanation of every score and formula\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n---\n\n## How It Works\n\n**9-step pipeline (Soft Predict — Claude runs all steps natively):**\n\n1. **Validate** — 5-rule check: entity real, system observable, time horizon set, signals available, question specific\n2. **Collect signals** — 6 web searches, minimum 18 signals, all classified by STEEEP + temporal + type\n3. **Score signals** — 4-factor formula: recency × reliability × type × evidence, regional multipliers applied\n4. **Extract drivers** — top 3 structural forces behind the signal clusters, ranked by score sum\n5. **Build STEEEP matrix** — 18-cell grid: 6 categories × 3 time horizons\n6. **Cross-impact analysis** — convergence and friction points across temporal layers\n7. **Find analogues** — 3 real historical cases, similarity scored, mapped to drivers\n8. **Compute predictions + confidence** — 3 independent scores (0–100 each, do NOT sum to 100); confidence penalizes blind spots\n9. **Write scenarios + report** — PROBABLE / PLAUSIBLE / POSSIBLE + PREFERABLE with IFTF backcasting\n\n**Hard Predict** extends to 12 steps with Python handling steps 1, 3, 5, 8, 9, 10, 12 deterministically (confidence, decision guidance, and report formatting each get their own dedicated step).\n\n---\n\n## IFTF Methodology\n\nThis plugin implements the [Institute for the Future](https://www.iftf.org) futures research framework:\n\n| IFTF Concept | Implementation |\n|---|---|\n| Futures Cone | PROBABLE / PLAUSIBLE / POSSIBLE / PREFERABLE |\n| Three Horizons | Operational (0–3yr) / Strategic (3–10yr) / Civilizational (10+yr) |\n| STEEEP Scan | 6-category signal collection and matrix |\n| Signals → Drivers | Step 4: extract structural forces from signal clusters |\n| Backcasting | PREFERABLE scenario works backwards from desired future |\n| Action Implications | DECISION GUIDANCE: stance, low-regret move, risk trigger |\n\n---\n\n## License\n\nMIT\n",
  "bytes": 5877,
  "sha": "303908e5c5ea1198b42e8d21d35b06346b03a383428dd564db3d756a1984a1b6",
  "repo_slug": "isanthoshgandhi/foresight-engine",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_isanthoshgandhi_foresight_engine_foresig_349ead0d/readme"
}