Brave Search
badlogic/pi-skills
Web search and content extraction via Brave Search API. An agent skill from badlogic/pi-skills.
Activate this skill for ANY future-oriented question. An agent skill from davepoon/buildwithclaude.
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install davepoon/buildwithclaude soft-predict-future --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .claude/skills/soft-predict-future && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .claude/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-futureType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install davepoon/buildwithclaude soft-predict-future --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .agents/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .agents/skills/soft-predict-future && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .agents/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install davepoon/buildwithclaude soft-predict-future --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .cursor/skills/soft-predict-future && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .cursor/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/davepoon/buildwithclaude.git --path plugins/foresight-intelligence/skills/soft-predict-future--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install davepoon/buildwithclaude soft-predict-future --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .gemini/skills/soft-predict-future && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .gemini/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install davepoon/buildwithclaude soft-predict-futureInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .github/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .github/skills/soft-predict-future && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .github/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add davepoon/buildwithclaude --skill soft-predict-future -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install davepoon/buildwithclaude soft-predict-future --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/davepoon/buildwithclaude.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/plugins/foresight-intelligence/skills/soft-predict-future .opencode/skills/soft-predict-future && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "soft-predict-future" agent skill from https://github.com/davepoon/buildwithclaude/tree/main/plugins/foresight-intelligence/skills/soft-predict-future into .opencode/skills/soft-predict-future/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "soft-predict-future", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
soft-predict-futureActivate this skill for ANY future-oriented question. An agent skill from davepoon/buildwithclaude.
Soft Predict Future is an agent skill from davepoon/buildwithclaude. Activate this skill for ANY future-oriented question. Triggers include: "Will [X]?", "Who will win [X]?", "What happens to [X]?", "Can [X] succeed?", "What's the future of X?", foresight analysis, scenario planning, STEEEP analysis, futures cone, prediction requests, or any question about a future outcome. Year is NOT required — the engine infers the horizon. Also activate when the user says "predict", "forecast", "what are the odds", "scenario analysis", or asks about competitive races, technology adoption…
Its SKILL.md is about 5.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Productivity & Automation, covering Web search. The repository describes itself as: A single hub to find Claude Skills, Agents, Commands, Hooks, Plugins, and Marketplace collections to extend Claude Code, Claude Desktop, Agent SDK and OpenClaw. The licence is MIT.
9 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 10bfc43. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Soft Predict Future loads about 5.4k tokens when it runs. Until then it costs about 169 tokens; SKILL.md has 1,302 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from davepoon/buildwithclaude at commit 10bfc43, republished under its MIT licence (© davepoon). 1,302 words, ~5,412 tokens.
.claude/skills/soft-predict-future/SKILL.md (or your agent's skills folder).Activate when the user asks any future-oriented question — "Will [X]?", "Who will win [X]?", "What happens to [X]?", "Can [X] succeed?", or any question about a future outcome. Year is NOT required. Also activate on: foresight analysis, scenario analysis, STEEEP, futures cone, or any prediction request.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
TRY ASKING (year optional — engine infers the horizon)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
■ Who will win — Google or Perplexity?
■ Will OpenAI or Anthropic dominate the AI race?
■ Will India become the global AI leader?
■ Will crypto replace banks?
■ Will remote work become permanent?
■ Will EVs dominate Indian cities by 2032?
■ Will UPI become Southeast Asia's default payment rail by 2028?
■ Will Europe lead the green energy transition by 2035?
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Soft Predict Future uses Claude's native reasoning + web search. Outputs are structurally correct and fast. For deterministic, auditable scoring say: "Run hard predict future: [your question]"
Execute ALL steps in order. Never skip. Never combine. Show your work at each step.
Apply exactly 5 binary rules. If ANY rule fails, stop and explain why. Do not proceed.
Rule 1 — Entity Reality: Does the entity actually exist in the real world? Fail if fictional, hypothetical, or unnamed.
Rule 2 — System Existence: Is the domain observable and researchable? Fail if purely philosophical or metaphysical.
Rule 3 — Time Horizon: Is the outcome observable within a 2–30 year window? A specific year is NOT required. If no year is given, infer the most reasonable horizon from the question's nature:
Fail ONLY if the implied timescale is geological, post-human, or clearly beyond 30 years.
After applying Rule 3, state the inferred horizon (e.g. "2026–2033" or "2028–2038").
Rule 4 — Signal Availability: Could real-world evidence plausibly exist? Fail if classified, purely speculative, or unpublished.
Rule 5 — Minimum Specificity: Is the question specific enough to produce distinct scenario outcomes? Fail if trivially true for any answer.
Output:
VALIDATION
Rule 1 Entity Reality: PASS / FAIL — [reason]
Rule 2 System Existence: PASS / FAIL — [reason]
Rule 3 Time Horizon: PASS / FAIL — [reason] | Inferred horizon: [YYYY–YYYY]
Rule 4 Signal Availability: PASS / FAIL — [reason]
Rule 5 Specificity: PASS / FAIL — [reason]
Result: PROCEED / STOPRun exactly 6 web searches. Collect a minimum of 18 signals total. Do not proceed with fewer than 18.
Search 1: Current state — "[topic] current status [year]"
Search 2: Growth indicators — "[topic] growth data market size [year]"
Search 3: Barriers and headwinds — "[topic] challenges barriers risks"
Search 4: Policy and regulation — "[topic] government policy regulation"
Search 5: Technology or infrastructure enablers — "[topic] technology infrastructure investment"
Search 6: Historical precedent — "[topic] historical analogue similar transition"
For each signal, classify all 6 attributes:
| Attribute | Values |
|---|---|
| direction | supporting / opposing / wildcard / neutral |
| steeep_category | Social / Technological / Economic / Environmental / Ethical / Political |
| temporal_layer | Operational (0–3yr) / Strategic (3–10yr) / Civilizational (10+yr) |
| source_type | primary / secondary / opinion |
| recency_days | integer |
| has_evidence | true / false (contains a number, date, or measurable fact) |
Present signals in a table with all 6 columns filled for every row.
Score every signal individually using this formula:
score = recency_weight × reliability_weight × type_weight × evidence_multiplierCap at 1.0. Round to 2 decimal places.
Recency weights:
Reliability weights:
Type weights:
Evidence multiplier:
Apply regional multiplier after base score:
final_score = min(1.0, base_score × regional_multiplier[steeep][temporal])Show scoring table: Signal | recency_w | reliability_w | type_w | evidence_mult | base_score | regional_mult | final_score
A driver is the deep structural force that explains WHY a cluster of signals exists. Signals are observable. Drivers are causal.
After scoring, group signals by STEEEP category. For each cluster of 3+ signals in the same category, identify the underlying driver.
Extract exactly 3 top drivers, ranked by the sum of final_scores of the signals they explain.
For each driver state:
Output:
STRUCTURAL DRIVERS
D1 [Name] — [Force]
Explains: [signal list] | Temporal: [layer] | Stability: [tier]
D2 [Name] — [Force]
Explains: [signal list] | Temporal: [layer] | Stability: [tier]
D3 [Name] — [Force]
Explains: [signal list] | Temporal: [layer] | Stability: [tier]Drivers feed directly into scenario writing in Step 8. Each scenario must be traceable to at least one driver.
Populate all 18 cells. Each cell value = average final_score of all signals mapped to that STEEEP × Temporal combination. Empty cells = 0.
| Operational (0–3yr) | Strategic (3–10yr) | Civilizational (10+yr) | |
|---|---|---|---|
| Social | |||
| Technological | |||
| Economic | |||
| Environmental | |||
| Ethical | |||
| Political |
Apply regional multipliers to each cell. Then identify:
Signals are scored independently in Step 3, but structural forces interact. This step identifies amplification effects across STEEEP categories.
Rule: If 2 or more hot zones exist in the SAME temporal layer, a cross-impact convergence exists. Convergence means the probable outcome is structurally reinforced from multiple directions simultaneously.
For each temporal layer (Operational / Strategic / Civilizational):
Also identify any opposing cross-impacts: where a hot zone in one STEEEP category directly contradicts or slows a hot zone in another (e.g. Technological/Strategic hot but Political/Strategic opposing). Flag these as FRICTION POINTS.
Output:
CROSS-IMPACT
Operational: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Strategic: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Civilizational: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Friction points: [list any STEEEP pairs in conflict, or "None detected"]
Convergence bonus: [+X% to probable_pct if Strategic convergence exists]Apply convergence bonus: if Strategic layer has CONVERGENCE, add 5% to probable_pct before normalization in Step 7.
Identify exactly 3 real historical cases that parallel the question's trajectory.
For each:
Prefer analogues with similarity ≥ 60%. If none exceed 60%, note as a confidence penalty.
Each future type is scored independently (0–100). They do NOT sum to 100%. Futures cone methodology: a scenario can be 80% Probable AND 60% Plausible simultaneously.
R_probable = (supporting signals with score > 0.70) × 3
+ (best analogue similarity / 100) × 4
+ (hot zone count) × 2
+ convergence_bonus (5 if Strategic CONVERGENCE, else 0)
R_plausible = (supporting signals with score 0.40–0.70) × 2
+ (second analogue similarity / 100) × 3
R_possible = (wildcard signals) × 2
+ (opposing signals with score > 0.60) × 2
+ (gap zones / 18) × 3Convert to independent scores (exponential curve, not normalization):
probable_score = min(100, round((1 - e^(-R_probable / 18)) × 100))
plausible_score = min(100, round((1 - e^(-R_plausible / 9)) × 100))
possible_score = min(100, round((1 - e^(-R_possible / 5)) × 100))signal_count_score = min(100, total_signals / 25 × 100) × 0.30
signal_diversity = (unique STEEEP categories covered / 6 × 100) × 0.30
recency_score = (signals with recency_days ≤ 90 / total) × 100 × 0.20
evidence_score = (signals with has_evidence=true / total) × 100 × 0.20
confidence = round(signal_count_score + signal_diversity + recency_score + evidence_score)PROBABLE, PLAUSIBLE, POSSIBLE — each must:
PREFERABLE — IFTF Backcasting Structure
Do not write PREFERABLE as a probability-weighted outcome. Write it as a designed future, then backcast to today.
Format:
■ PREFERABLE — [Short title]
[2–3 sentences: describe the desired state as already achieved]
BACKCAST
Civilizational (10+yr): [What must be structurally true by the far horizon]
Strategic (3–10yr): [What must be built or decided in the medium horizon]
Operational (0–3yr): [What must happen NOW to set the trajectory]
LEVERAGE: [The single highest-leverage action available today — specific, not generic]
DRIVER: [Which structural driver (D1/D2/D3) this path depends on most]IF probable_score > 60:
stance = "Align with probable scenario trajectory"
low_regret = "Invest in capability building in the dominant hot zone"
ELIF plausible_score > 50:
stance = "Hedge between probable and plausible scenarios"
low_regret = "Choose reversible commitments that work in both"
ELIF possible_score > 40:
stance = "Maintain optionality — signal environment is ambiguous"
low_regret = "Invest in monitoring and early-warning indicators"
ELSE:
stance = "Defer commitment — insufficient signal clarity"
low_regret = "Reduce uncertainty before acting"Confidence qualifier:
Risk trigger: the opposing signal with the highest final_score.
MANDATORY: Output ALL sections below, every single run, no exceptions. Never skip a section. Never produce a partial report.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
SOFT PREDICT FUTURE · FORESIGHT ENGINE
[Query]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
PREDICTIONS
■ Probable [[X]/100] [████████████░░░░░░░░] — [one sentence, no hedging]
■ Plausible [[X]/100] [████████░░░░░░░░░░░░] — [one sentence, no hedging]
■ Possible [[X]/100] [████░░░░░░░░░░░░░░░░] — [one sentence, no hedging]
■ Preferable [stakeholder analysis below]
Confidence: [X]/100 | Signals: [N] | Horizon: [YYYY–YYYY] | [YYYY-MM-DD]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
SIGNAL PULSE
— How many pieces of real-world evidence support, oppose, or complicate this question
Supporting [N] [████████████░░░░░░░░] Opposing [N] [████░░░░░░░░░░░░░░░░] Wild [N]
Net: [SUPPORTING LEADS / OPPOSING LEADS / NEUTRAL]
Hot zone: [The single STEEEP category with strongest evidence]
Gap: [STEEEP categories with no signals, or "None — full coverage"]
STRUCTURAL DRIVERS
— The 3 deep causal forces (not events) explaining WHY the signals exist. Stability = likelihood of change.
D1 [Name] — [Force] ([LOCKED / SHIFTING / FRAGILE])
D2 [Name] — [Force] ([LOCKED / SHIFTING / FRAGILE])
D3 [Name] — [Force] ([LOCKED / SHIFTING / FRAGILE])
CROSS-IMPACT
— Whether multiple STEEEP domains reinforce or contradict each other in the same time layer
Operational: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Strategic: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Civilizational: [CONVERGENCE / ISOLATED / BLIND LAYER] — [explanation]
Friction: [Conflicting domain pairs, or "None detected"]
HISTORICAL MATCH
— Real past transition most structurally similar to this question. Higher % = stronger precedent.
[Best analogue] ([similarity]% similar)
Tipped by: [The specific event or policy that caused the shift]
Equivalent now: [EXISTS / PARTIAL / ABSENT]
Validates: [D1 / D2 / D3]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
■ PROBABLE [[X]%] — [Title]
— The most evidence-backed outcome given current signal strength
[2–3 sentence narrative. No hedging.]
PROOF: [Fact with number or date]
IF: [The condition that activates this scenario]
BUT: [The constraint or bottleneck that could slow it]
DRIVER: D[n]
■ PLAUSIBLE [[X]%] — [Title]
— A realistic alternative if moderate signals strengthen or dominant ones weaken
[2–3 sentence narrative]
PROOF: [Fact with number or date]
IF: [Activation condition]
BUT: [Constraint]
DRIVER: D[n]
■ POSSIBLE [[X]%] — [Title]
— A lower-probability outcome driven by wildcards or high-scoring opposing signals
[2–3 sentence narrative]
PROOF: [Fact with number or date]
IF: [Activation condition]
BUT: [Constraint]
DRIVER: D[n]
■ PREFERABLE — [Title]
— Not a prediction. A designed future: what the best achievable outcome looks like, traced back to today.
[2–3 sentences: desired state as already achieved. No hedging.]
BACKCAST
Civilizational: [What must be structurally true by the far horizon]
Strategic: [What must be built or decided in the medium term]
Operational: [What must begin NOW to set the trajectory]
LEVERAGE: [Single highest-leverage action today — specific actor, specific action]
DRIVER: D[n]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
PREFERABLE FUTURES · Per major stakeholder
— For each major player in the query, state the conditions required for their preferred outcome
[Player A]:
Wins IF → [specific condition that must be created or occur]
BUT ONLY → [binding constraint that must also be satisfied]
ONLY THEN → [the outcome that becomes possible]
[Player B]:
Wins IF → [specific condition]
BUT ONLY → [binding constraint]
ONLY THEN → [outcome]
[Users/Society — always include]:
Wins IF → [condition for best collective outcome]
BUT ONLY → [constraint]
ONLY THEN → [outcome]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
THE ONE THING
— The single variable whose presence or absence determines which scenario actually unfolds
[One sentence naming the deciding variable]
INCIDENT: [Real past event showing this variable's power]
WATCH: [Leading indicator — a milestone, metric, or policy action to monitor]
IF YES → [What accelerates]
IF NO → [What stalls]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
DECISION GUIDANCE
Recommended stance: [From deterministic probability logic]
Low-regret move: [Action that pays off in multiple scenarios simultaneously]
Risk trigger: [Highest-scored opposing signal — the one that could invalidate Probable]
[REGIONAL LENS — [REGION]]
Top multipliers: [STEEEP/temporal (Xx)] [STEEEP/temporal (Xx)]
Key local variable: [One sentence on the dominant local structural factor]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
METHODOLOGY KEY
Signal score (0–1) Recency × source reliability × signal type × evidence strength — higher = fresher, better-sourced, stronger evidence
Confidence (0–100) Signal density (0–40) + evidence balance (0–30) + historical grounding (0–30) − blind spot penalty (0–15)
Predictions PROBABLE / PLAUSIBLE / POSSIBLE are independent scores (0–100 each, do NOT sum to 100)
Futures cone: a scenario can score high on multiple types simultaneously
STEEEP matrix 6 domains × 3 time horizons — ★ hot (>1.0) ● warm (>0.5) ✗ blind spot (0)
Historical similarity Structural pattern match to real past transitions — 60%+ is reliable precedent; below 40% is weak grounding
Convergence bonus +5 added to Probable score when 2+ STEEEP domains reinforce each other in the Strategic layer
Stability tiers LOCKED = unlikely to change in 10yr | SHIFTING = could change in 3–5yr | FRAGILE = could reverse in 1–2yr
Preferable futures Per stakeholder: Wins IF [condition] BUT ONLY [constraint] ONLY THEN [outcome]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
---
## Visual Output (claude.ai with Artifacts)
If running on claude.ai and Artifacts are enabled, after the text report generate an HTML Artifact:
```html
<!-- Render a visual foresight report with:
1. Predictions bar chart — horizontal bars for Probable/Plausible/Possible scores
2. STEEEP matrix — 6×3 color-coded table (darker green = hotter cell, red = blind spot)
3. Futures cone — SVG diagram showing 4 scenario bands expanding from present to horizon
4. Stakeholder preferable cards — one card per player with Wins IF / BUT ONLY / ONLY THEN
Use inline CSS only. No external dependencies. Dark background (#0f0f0f), accent color #00d4aa.
-->Apply in Step 3 and Step 5.
| Operational | Strategic | Civilizational | |
|---|---|---|---|
| Social | 1.10 | 1.30 | 1.20 |
| Technological | 1.40 | 1.30 | 1.10 |
| Economic | 1.20 | 1.25 | 1.15 |
| Environmental | 0.90 | 1.00 | 1.10 |
| Ethical | 0.95 | 1.00 | 1.05 |
| Political | 0.85 | 0.90 | 1.00 |
India note: UPI/DPI gives asymmetric advantage in Technological/Operational. Political/Operational discounted by regulatory fragmentation across states.
| Operational | Strategic | Civilizational | |
|---|---|---|---|
| Social | 1.00 | 1.10 | 1.05 |
| Technological | 1.20 | 1.40 | 1.20 |
| Economic | 1.10 | 1.30 | 1.10 |
| Environmental | 0.95 | 1.00 | 1.05 |
| Ethical | 1.05 | 1.10 | 1.10 |
| Political | 0.90 | 0.95 | 1.00 |
USA note: Deep capital markets amplify Technological/Strategic. Political/Operational discounted by legislative gridlock.
| Operational | Strategic | Civilizational | |
|---|---|---|---|
| Social | 1.00 | 1.05 | 1.10 |
| Technological | 1.00 | 1.10 | 1.05 |
| Economic | 0.95 | 0.90 | 0.90 |
| Environmental | 1.20 | 1.40 | 1.30 |
| Ethical | 1.10 | 1.20 | 1.20 |
| Political | 1.05 | 1.10 | 1.10 |
Europe note: Regulatory leadership (GDPR, EU AI Act, Green Deal) amplifies Environmental/Strategic. Economic/Civilizational discounted by demographic headwinds.
| Operational | Strategic | Civilizational | |
|---|---|---|---|
| Social | 1.00 | 1.10 | 1.05 |
| Technological | 1.20 | 1.50 | 1.30 |
| Economic | 1.10 | 1.20 | 1.10 |
| Environmental | 0.90 | 1.00 | 1.05 |
| Ethical | 0.70 | 0.75 | 0.80 |
| Political | 1.10 | 1.15 | 1.00 |
China note: State-directed capital amplifies Technological/Strategic strongly. Ethical/Operational discounted by limited transparency.
All multipliers = 1.0. Apply when no region is detectable.
© davepoon, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Just SKILL.md in plugins/foresight-intelligence/skills/soft-predict-future of davepoon/buildwithclaude.
Open the folder on GitHubat commit 10bfc43
We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in davepoon/buildwithclaude, which our catalogue first saw on October 7, 2026.
Soft Predict Future next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Soft Predict Future this skilldavepoon/buildwithclaude | 3.6k | 1 repos | ~5.4k | Automated safety check: Pass | MIT | |
| Brave Searchbadlogic/pi-skills | 2.6k | 6 repos | ~592 | Automated safety check: Pass | MIT | |
| Enterprise AI Scenario MapMetaInFLow/Enterprise-ai-scenario-map-skill | 632 | — | ~1.8k | Automated safety check: Pass | MIT | |
| Web Searchjjyaoao/HelloAgents | 3.2k | 1 repos | ~5.6k | Automated safety check: Pass | MIT | |
| Ddg SearchTheSyart/claude-agent-examples | 405 | 1 repos | ~493 | Automated safety check: Pass | None | |
| Local Web SearchuluckyXH/OpenMOSS | 1.3k | — | ~392 | Automated safety check: Notes | MIT |
badlogic/pi-skills
Web search and content extraction via Brave Search API. An agent skill from badlogic/pi-skills.
MetaInFLow/Enterprise-ai-scenario-map-skill
企业AI场景地图生成报告工具。通过 web-search 深度调研企业信息,按照V2.1标准模板生成结构化AI应用场景地图报告,包含企业画像、业务诊断、行业实践、AI场景全量表、实施路径等完整内容。
jjyaoao/HelloAgents
Implement web search capabilities using the z-ai-web-dev-sdk.
TheSyart/claude-agent-examples
Web search without an API key using DuckDuckGo Lite via webfetch.
uluckyXH/OpenMOSS
A skill your agent uses when the user asks for web search that should run via the local-160 Responses API with websearch tool (base URL like https://proxy.example.com, model gpt-5.2-codex(xhigh)).
EXboys/skilllite
Web search and content extraction with Tavily and Exa via inference.sh CLI.
davepoon/buildwithclaude
A skill your agent uses when the user asks to "analyze video", "watch this video", "what happens in this video", "describe this clip", "review this footage", "classify these videos", "compare…
davepoon/buildwithclaude
Activate this agent for any future-oriented question that requires deep quantitative analysis, historical precedents, and structured scenario planning.
davepoon/buildwithclaude
Build, update, and apply iOS design specifications using Apple Human Interface Guidelines (HIG) source data.
davepoon/buildwithclaude
Download YouTube videos with customizable quality and format options.
davepoon/buildwithclaude
Discover Atlas Cloud image and video models, inspect their live schemas, and submit one confirmed media generation request with bounded GET polling.
davepoon/buildwithclaude
Toolkit for creating animated GIFs optimized for Slack, with validators for size constraints and composable animation primitives.
Categories
Activate this skill for ANY future-oriented question. An agent skill from davepoon/buildwithclaude. Soft Predict Future is an agent skill from davepoon/buildwithclaude. Activate this skill for ANY future-oriented question.
Soft Predict Future fits situations like: include: Will [X]?; who will win [X]?; what happens to [X]?; can [X] succeed?.
Run `npx skills add davepoon/buildwithclaude --skill soft-predict-future -a claude-code`. Or copy the skill folder (plugins/foresight-intelligence/skills/soft-predict-future in davepoon/buildwithclaude) into .claude/skills/soft-predict-future in your project. Claude Code loads it when a task matches its description.
Run `npx skills add davepoon/buildwithclaude --skill soft-predict-future -a codex`. Or copy the skill folder (plugins/foresight-intelligence/skills/soft-predict-future in davepoon/buildwithclaude) into .agents/skills/soft-predict-future in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add davepoon/buildwithclaude --skill soft-predict-future -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/soft-predict-future, .gemini/skills/soft-predict-future, .github/skills/soft-predict-future and .opencode/skills/soft-predict-future in your project.
SKILL.md names no scripts, command-line tools or credentials: Soft Predict Future is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Soft Predict Future is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 5.4k tokens (SKILL.md is roughly 22k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Soft Predict Future: Brave Search (badlogic/pi-skills, 2.6k stars), Enterprise AI Scenario Map (MetaInFLow/Enterprise-ai-scenario-map-skill, 632 stars), Web Search (jjyaoao/HelloAgents, 3.2k stars) and Ddg Search (TheSyart/claude-agent-examples, 405 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
davepoon (a GitHub user) maintains it in davepoon/buildwithclaude, which has 3,604 GitHub stars. The repository holds 245 skills in this directory. The repository was last updated on October 6, 2026.
Source: davepoon/buildwithclaude on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.