Purpose
Produce a TCFD-aligned climate risk register that names the specific physical and transition risks a business faces, scores each by likelihood and financial impact, and pairs every material risk with a concrete mitigation. Replaces vague "we should think about climate" conversations with a defensible, board-ready risk inventory.
When to Use
Invoke this skill when the user asks to:
- Conduct a climate risk assessment for a company, division, or asset portfolio
- Build a climate risk register aligned to TCFD
- Prepare physical and transition risk disclosures for an annual report or ESG filing
- Quantify financial exposure to climate scenarios (e.g., 1.5°C, 2°C, 4°C pathways)
- Identify which climate risks are material and which mitigations to prioritize
- Brief a board or audit committee on climate risk exposure
Inputs Required
Before starting, collect:
- Industry/sector: The primary business activity (e.g., food manufacturing, real estate, logistics)
- Key geographies: Where operations, facilities, and major suppliers are located
- Asset base: Owned facilities, leased sites, critical infrastructure, or asset-light model
- Supply chain exposure: Key inputs, single-source suppliers, and high-risk sourcing regions
- Revenue model: How the business makes money and which products/services drive it
- Time horizons: Definition of short (e.g., 0–3 yrs), medium (3–10 yrs), long-term (10+ yrs) — or use defaults
- Regulatory context: Carbon pricing exposure, disclosure mandates, or jurisdiction-specific rules
If time horizons are not provided, use short (0–3 years), medium (3–10 years), long (10–30 years). If a specific scenario isn't named, assess against both a 2°C orderly transition and a 4°C high-physical-risk pathway.
Instructions
Step 1: Establish Scope and Scenario Frame
Anchor the assessment before scoring anything:
- Confirm the organizational boundary (operations, value chain, or both)
- Set the two scenario lenses: a transition-heavy scenario (rapid decarbonization, ~2°C — high policy/market/tech risk) and a physical-heavy scenario (limited action, ~4°C — high acute/chronic risk)
- Map the time horizons to the business's planning and investment cycles
Step 2: Identify Physical Risks
Catalog risks tied to the physical effects of climate change, split into two types.
Acute (event-driven) risks — assess relevance of each to the asset base and geographies:
- Flooding (river, coastal, surface water)
- Tropical cyclones / severe storms / wind damage
- Wildfire
- Extreme heat events
- Drought-driven water scarcity
Chronic (long-term shift) risks:
- Rising mean temperatures affecting operations, labor, or yields
- Sea-level rise threatening coastal assets
- Changing precipitation patterns and water availability
- Permanent shifts in growing seasons or resource viability
For each physical risk flagged as relevant, note: the exposed asset/operation, the geography, the affected time horizon, and the transmission channel to the business (e.g., production halt, asset damage, supply disruption).
Step 3: Identify Transition Risks
Catalog risks from the shift to a lower-carbon economy across four TCFD categories.
Policy & Legal: Carbon pricing/taxes, emissions caps, disclosure mandates, litigation exposure, building/efficiency standards.
Technology: Stranded assets from low-carbon substitution, cost of transitioning processes, failed technology bets.
Market: Shifting customer demand, rising cost or scarcity of carbon-intensive inputs, changing energy prices.
Reputation: Stakeholder and investor pressure, brand exposure on emissions, sector stigmatization.
For each transition risk flagged as relevant, note: the driver, the time horizon when it bites, and the transmission channel (cost increase, revenue loss, asset write-down, cost of capital).
Step 4: Score Each Risk
Apply a consistent two-dimensional scoring model to every identified risk.
Likelihood (within the relevant time horizon and scenario):
- Rare / Unlikely / Possible / Likely / Almost Certain (1–5)
Financial impact (against revenue, EBITDA, or asset value — state the basis):
- Insignificant / Minor / Moderate / Major / Severe (1–5)
- Where possible, attach a rough quantified range or % of revenue/EBITDA exposed
Risk rating = Likelihood × Impact, banded:
- 1–4 Low / 5–9 Medium / 10–15 High / 16–25 Critical
Note the velocity (how fast the risk materializes) for any High or Critical risk, since slow-onset risks allow more mitigation runway.
Step 5: Build the Risk Register
Compile all scored risks into a single register, sorted by risk rating (Critical first). Each row includes:
- Risk ID and name
- Type (Physical-Acute / Physical-Chronic / Transition-Policy / -Technology / -Market / -Reputation)
- Scenario and time horizon
- Likelihood (1–5), Financial impact (1–5), Risk rating + band
- Transmission channel to financials
- Mitigation recommendation
Step 6: Recommend Mitigations
For each High and Critical risk, provide a specific, actionable mitigation:
- Name the action (e.g., relocate distribution center out of 1-in-100 floodplain; hedge carbon-input cost; diversify single-source supplier in drought region)
- State the risk-reduction lever (avoid / reduce / transfer / accept)
- Note rough effort/cost level and the residual risk after mitigation
Step 7: Summarize Top Exposures
Close with a board-ready summary: the 3–5 most material risks, total estimated exposure if quantifiable, and the top 3 mitigation priorities to fund first.
Output Format
- Structure: Scope/scenario frame → physical risk inventory → transition risk inventory → scoring methodology note → full risk register (table) → mitigation detail for High/Critical risks → executive summary
- Length: 1,200–1,800 words plus the register table
- Format: Structured document with one register table; mitigations and summary in structured bullets
- Sections: Scope & Scenarios → Physical Risks → Transition Risks → Scoring Methodology → Risk Register → Mitigation Recommendations → Top Exposures Summary
- Tone: Risk-committee analytical — precise, defensible, no alarmism and no greenwashing; every claim tied to a transmission channel
Quality Checklist
Before delivering output, verify:
- [ ] Both physical (acute + chronic) and transition (all 4 TCFD categories) risks are covered — none of the six types is empty without justification
- [ ] Every risk in the register has a likelihood score, an impact score, and a calculated risk rating + band
- [ ] Each risk states a specific transmission channel to financials — not a generic "could be costly"
- [ ] The register is sorted by risk rating with Critical/High risks at the top
- [ ] Every High and Critical risk has a named mitigation with a risk-reduction lever and residual risk
- [ ] Risks are tied to the company's actual geographies, assets, and supply chain — not a generic sector list
- [ ] The executive summary names a specific 3–5 most material risks and top 3 funding priorities
Notes
- Physical risk concentration usually hides in the supply chain, not the company's own footprint — probe single-source suppliers in high-exposure regions
- Transition risk timing is the hard part: a risk can be "almost certain" in the long term but "rare" in the short term — always score against the stated horizon
- If financial figures aren't available, score impact qualitatively but flag which risks most warrant a quantified follow-up study
- This assessment supports TCFD's "Risk Management" and "Strategy" pillars; pair it with scenario analysis and metrics/targets work for full disclosure readiness
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills