back
loading skill details...
Assess physical and transition climate risk for a site, product, or portfolio with scenario-based structure. Use when asked to run a climate risk assessment,...
---
name: climate-risk-assessment
description: "Assess physical and transition climate risk for a site, product, or portfolio with scenario-based structure. Use when asked to run a climate risk assessment, evaluate physical or transition risk, prepare TCFD/ESRS-style climate risk analysis, or assess how climate scenarios affect an asset or business. Produces a hazard-exposure-vulnerability matrix across 2030/2040/2050 horizons and labelled scenarios, with financial-impact ranges, confidence levels, and adaptation options."
homepage: https://mohitagw15856.github.io/pm-claude-skills/skill/climate-risk-assessment.html
metadata:
{
"openclaw": { "emoji": "🌍" }
}
---
# Climate Risk Assessment Skill
Climate risk work goes wrong by being either vague ("weather may worsen") or falsely precise ("flood losses of $4.7M in 2043"). This skill structures an assessment as hazard × exposure × vulnerability, across explicit time horizons and named scenarios, with impact expressed as ranges plus a confidence label. It supports decision-making; it is not an engineering or actuarial study.
## What This Skill Produces
- A risk register for the asset(s): hazard × exposure × vulnerability, scored per horizon and scenario
- Physical risks (acute + chronic) and transition risks (policy, market, technology, reputation) treated separately
- Financial-impact ranges with an explicit confidence level per estimate
- Adaptation and mitigation options with rough cost/benefit direction
- A stated-assumptions register so the assessment can be challenged and updated
## Required Inputs
Ask for these if not provided; with a thin brief, proceed using labelled assumptions (e.g. "assumed coastal location — confirm") rather than refusing:
- **Subject** — the site(s), product, or portfolio, with location(s) and asset life / planning horizon
- **Exposure basics** — what is physically and economically at stake: asset value, revenue dependence, supply chain nodes, workforce
- **Known sensitivities** — past climate-related disruptions, insurance history, single points of failure
- **Transition context** — sector, carbon intensity, regulatory exposure, customer decarbonization pressure
- **Financial scale** — revenue or asset value bands, so impact ranges land in the right order of magnitude
## Assessment Framework
**1. Structure every risk as hazard × exposure × vulnerability.** A hazard (e.g. riverine flood) only becomes a risk when exposure (asset in the floodplain) meets vulnerability (no defenses, 6-week recovery). Score each component, not just the headline.
**2. Use three horizons: 2030 / 2040 / 2050.** Near-term physical risk is mostly locked in regardless of scenario; scenario divergence dominates from ~2040. Say which effect drives each rating.
**3. Label scenarios and use them consistently:**
| Label | Character | Physical risk | Transition risk |
|---|---|---|---|
| Orderly | Early, coordinated policy (~1.5–2°C) | Lower long-term | Front-loaded but predictable |
| Disorderly | Late, abrupt policy (~<2°C, delayed) | Moderate | Sharp, repricing shocks in 2030s |
| Hothouse | Weak policy (~3°C+) | High and compounding | Low policy risk, high physical + liability |
Assess at least orderly and hothouse — they bracket the outcome space. Note that transition and physical risk peak in *different* scenarios; a single-scenario view always understates one of them.
**4. Express impact as a range with confidence.** Use order-of-magnitude bands tied to the subject's financial scale (e.g. "1–5% of site revenue per event"), and tag each estimate High / Medium / Low confidence with the reason (data quality, model dependence, deep uncertainty).
**5. Identify adaptation options per material risk.** Categories: harden (physical protection), diversify (sites, suppliers), transfer (insurance, contracts), retreat (relocate, exit), and accept (with monitoring trigger). Note rough cost direction and by when a decision is needed.
## Output Format
### Climate risk assessment: [subject]
**1. Scope and assumptions** — subject, horizons, scenarios used, and every assumption made from a thin brief, labelled.
**2. Physical risk register** — table: hazard | type (acute/chronic) | exposure | vulnerability | rating per horizon × scenario | confidence.
**3. Transition risk register** — same structure across policy, market, technology, and reputation risks (include transition *opportunities* where real).
**4. Financial impact summary** — the top 5 risks with impact ranges, confidence, and which scenario/horizon drives each.
**5. Adaptation options** — per material risk: option, category, cost direction, decision-by date.
**6. Monitoring triggers** — observable signals (regulation drafts, insurance repricing, hazard events) that should force a reassessment.
Include this line in the artifact: *"Verify scenario selection, disclosure use, and any figures intended for reporting against the applicable standard/regulation (e.g. ESRS E1, TCFD/ISSB) with your compliance team; use engineering-grade studies before capital decisions."*
## Quality Checks
- [ ] Every risk decomposes into hazard, exposure, and vulnerability — no monolithic "climate risk" entries
- [ ] All three horizons and at least two labelled scenarios are covered, and the driver of each rating is stated
- [ ] Every financial estimate is a range with a confidence label and a reason for that confidence
- [ ] Transition and physical risks are assessed separately and both appear in the top-risks summary
- [ ] Each material risk has at least one adaptation option with a decision-by date
- [ ] Assumptions and monitoring triggers are explicit enough for a future reassessment
## Anti-Patterns
- [ ] Do not give point estimates for long-horizon losses — ranges with confidence, always
- [ ] Do not assess only one scenario — a single pathway hides either transition or physical risk
- [ ] Do not conflate hazard with risk — a flood map is not a loss estimate until exposure and vulnerability are in
- [ ] Do not let 2050 risks crowd out 2030 decisions — state what must be decided this planning cycle
- [ ] Do not present this as an engineering, actuarial, or compliance-grade study — it structures judgment, it does not replace specialist analysis
## Based On
TCFD/ISSB and ESRS E1 climate-risk practice (physical/transition split, scenario analysis, hazard–exposure–vulnerability structure).
don't have the plugin yet? install it then click "run inline in claude" again.
separated intent, inputs, procedure (8 numbered steps with explicit in/out), decision points (7 branches covering thin briefs, single scenarios, missing data, sector variability, disclosure intent), output contract (7 detailed sections with file and format specs), and outcome signal (7 observable criteria) from original monolithic guidance.
assess physical and transition climate risk for a site, product, or portfolio by decomposing each risk into hazard, exposure, and vulnerability components, then scoring them across three time horizons (2030/2040/2050) and named climate scenarios (orderly, disorderly, hothouse). use this skill when a user asks to run a climate risk assessment, evaluate physical or transition risk, prepare a TCFD/ESRS-style climate risk analysis, or determine how climate scenarios affect an asset or business unit. the output structures judgment to support strategic planning, not engineering or actuarial work. it prevents vague ("weather may worsen") and falsely precise ("exact $4.7M loss in 2043") assessments by forcing explicit assumptions, impact ranges tied to financial scale, and confidence labels on every estimate.
note on thin briefs: if subject, exposure, or transition context is incomplete, proceed with labelled assumptions (e.g. "assumed temperate coastal location, confirm") rather than refusing. ask clarifying questions in-line.
step 1: decompose each risk into hazard, exposure, vulnerability.
input: subject description and known sensitivities.
process: for each climate hazard relevant to the subject (e.g. riverine flood, coastal surge, drought, heatwave, policy tightening), identify (a) the hazard itself (e.g. 1-in-100-year flood depth), (b) exposure (asset or revenue in the affected area or supply chain), (c) vulnerability (adaptive capacity: defenses, insurance, recovery time, redundancy). do not score a monolithic "climate risk"; break it into three components.
output: a list of identified hazards with exposure and vulnerability sketched for each.
step 2: establish time horizons and scenario labels.
input: asset lifespan, planning cycle, disclosure requirement (if any).
process: commit to three horizons: 2030 (near-term, mostly locked-in physical risk), 2040 (peak scenario divergence), 2050 (long-term, high uncertainty). define three named scenarios:
assess at least orderly and hothouse; disorderly is optional but recommended. note that transition and physical risk peak in different scenarios; single-scenario analysis always understates one.
output: list of horizons and scenario definitions with key parameters (warming, policy timing, carbon price paths if quantified).
step 3: score physical risks (acute and chronic) per hazard, exposure, vulnerability.
input: identified hazards, exposure and vulnerability sketches, scenario definitions.
process: for each hazard:
output: physical-risk register table (hazard | type | exposure | vulnerability | 2030-orderly | 2030-disorderly | 2030-hothouse | 2040-orderly | 2040-disorderly | 2040-hothouse | 2050-orderly | 2050-disorderly | 2050-hothouse | driver).
step 4: score transition risks (policy, market, technology, reputation) per scenario.
input: transition context (sector, carbon intensity, regulatory exposure, customer pressure), scenario definitions.
process: assess separately across four transition-risk categories:
for each transition-risk category, assign a rating per scenario and horizon. note that orderly scenarios have higher transition risk upfront (early policy cost) but lower physical risk long-term; hothouse has lower early transition risk (no policy cost) but higher physical and liability risk.
include transition opportunities where material (e.g. "early adoption of low-carbon tech in orderly scenario reduces operating costs by 2040").
output: transition-risk register table (risk category | orderly 2030/2040/2050 | disorderly 2030/2040/2050 | hothouse 2030/2040/2050 | confidence | driver).
step 5: quantify financial impact ranges with confidence labels.
input: top physical and transition risks, financial scale (revenue or asset value), exposure and vulnerability scores.
process: for each material risk (top 5-10 by rating), estimate annual or event-based impact as a percentage or absolute range tied to the subject's financial scale.
output: financial-impact summary table (risk | scenario/horizon | impact range | confidence | reason).
step 6: identify adaptation and mitigation options per material risk.
input: physical-risk register, transition-risk register, financial impact summary.
process: for each material risk, propose 1-3 adaptation/mitigation options from these categories:
give rough cost direction (low/medium/high capex or opex, one-time vs. recurring) and a decision-by date tied to the planning horizon.
output: adaptation-options table (risk | option | category | cost direction | decision-by | rationale).
step 7: define monitoring triggers and document all assumptions.
input: all preceding outputs.
process:
output: monitoring-triggers list and assumptions register.
step 8: compile and review output.
input: all preceding tables, assumptions, triggers.
process: assemble the artifact (see output contract). run quality checks. include mandatory disclaimer: "Verify scenario selection, disclosure use, and any figures intended for reporting against the applicable standard/regulation (e.g. ESRS E1, TCFD/ISSB) with your compliance team; use engineering-grade studies before capital decisions."
output: complete climate-risk assessment artifact.
if subject or exposure is vague (e.g. "a manufacturing company"), proceed with labelled assumptions and ask for clarification in-line rather than refusing. example: "assumed temperate northern Europe location, single site, €50M revenue, confirm".
if only one scenario is provided or requested, insist on a second scenario (orderly + hothouse minimum) to avoid understating either transition or physical risk. explain that single-scenario analysis hides one dimension.
if financial scale is not given, estimate order-of-magnitude bands from context (e.g. if a small business, assume €0.5-5M revenue; if a large corp, assume €1B+). state the assumption and ask the user to correct it.
if the subject is outside high-risk sectors (e.g. tech, pharma in temperate zone with no supply-chain exposure), transition risk may dominate and physical risk be low. confirm scope with the user; do not force a lengthy physical-risk register if it is not material.
if deep uncertainty is high (e.g. emerging market, novel supply chain, nascent technology exposure), use low-confidence estimates with wide ranges and emphasize monitoring triggers and scenario flexibility over point forecasts.
if the user intends to disclose output (TCFD, ESRS, SEC, etc.), check regulatory definitions and add a note on alignment (e.g. "aligned with TCFD Task Force recommendations on governance, strategy, risk management, metrics"; or "mapped to ESRS E1 materiality thresholds").
if no adaptation options are obvious, default to "accept with monitoring triggers" and set a review cadence (e.g. "annual reassessment if no major events; ad-hoc if insurance repricing >10%").
produce a single artifact titled "Climate Risk Assessment: [subject]" containing these sections in order:
scope and assumptions , subject (site, location, asset life), horizons (2030/2040/2050), scenarios (orderly, disorderly, hothouse with warming and policy parameters), and a labelled assumptions register (every assumption from thin brief).
physical risk register , table with columns: hazard | type (acute/chronic) | exposure score | vulnerability score | 2030-orderly | 2030-disorderly | 2030-hothouse | 2040-orderly | 2040-disorderly | 2040-hothouse | 2050-orderly | 2050-disorderly | 2050-hothouse | driver (climate vs. vulnerability).
transition risk register , table with columns: risk category (policy, market, technology, reputation) | 2030-orderly | 2030-disorderly | 2030-hothouse | 2040-orderly | 2040-disorderly | 2040-hothouse | 2050-orderly | 2050-disorderly | 2050-hothouse | confidence | driver.
financial impact summary , top 5-10 risks with: risk name | scenario(s) driving impact | impact range (% of revenue or €M) | confidence (high/medium/low) | reason for confidence.
adaptation options , table with columns: risk | option | category (harden/diversify/transfer/retreat/accept) | cost direction | decision-by | rationale.
monitoring triggers , bulleted list of observable signals (regulation, insurance repricing, hazard events, tech breakthroughs, peer announcements) tied to specific risks and horizons; include a reassessment cadence (e.g. annual or event-driven).
mandatory disclaimer (at the end): "Verify scenario selection, disclosure use, and any figures intended for reporting against the applicable standard/regulation (e.g. ESRS E1, TCFD/ISSB) with your compliance team; use engineering-grade studies before capital decisions."
the user knows the skill worked when:
the assessment decomposes every risk into hazard, exposure, and vulnerability (not monolithic "climate risk" entries).
the user can state which horizon and scenario drive each risk rating (e.g. "physical flood risk peaks in 2050 hothouse because exposure stays high and adaptation is late").
every financial estimate is a range with a confidence label and a stated reason (e.g. "1-5% revenue loss per event, medium confidence because 30 years of flood data exists but climate shifts may change frequency").
transition and physical risks are visibly separated and both appear in the top-risks summary; the user notes that orderly and hothouse scenarios have opposite transition-vs-physical risk profiles.
each material risk has at least one adaptation option with a decision-by date (e.g. "harden defenses, capex, decide by 2026 before 2030 flood risk increases").
the assumptions register is explicit enough that a colleague or auditor can challenge and update specific entries without rerunning the whole assessment.
monitoring triggers are concrete and tied to specific risks (e.g. "if EU carbon price >€100/tonne by 2028, repricing in this assessment").
based on TCFD (Task Force on Climate-related Financial Disclosures), ISSB (International Sustainability Standards Board), and ESRS E1 (European Sustainability Reporting Standard on climate change) physical/transition split, scenario analysis, and hazard-exposure-vulnerability structure.
original skill author: clawhub; enriched for Implexa standards.