Purpose
Convert a described climate risk into a defensible, standardized materiality rating — scoring likelihood against financial impact across three time horizons using the TCFD taxonomy — so risk teams can prioritize, register, and disclose climate exposures consistently rather than scoring each risk ad hoc.
When to Use
Invoke this skill when the user asks to:
- Score or rate the materiality of a specific climate risk
- Classify a risk under the TCFD physical/transition taxonomy
- Assess how a risk's severity changes across short, medium, and long-term horizons
- Build or populate a climate risk register with consistent scoring
- Prioritize climate risks for TCFD, ISSB (IFRS S2), or CSRD disclosure
- Compare a physical risk against a transition risk on a common scale
Inputs Required
Before starting, collect:
- Risk description: 2–4 sentence description of the climate risk being assessed
- Organization context: Industry, geography of operations/assets, and rough revenue scale
- Exposed assets or activities: What is at risk — facilities, supply chain, revenue lines, regulatory obligations
- Existing controls/mitigations: Adaptation measures, insurance, hedges, or none
- Time horizon definitions: How the organization defines short/medium/long term (default: short ≤2y, medium 2–10y, long 10–30y)
- Financial materiality threshold: Revenue or EBITDA percentage that counts as material (default: 1% revenue = low band, 5% = high band)
- Scenario assumption: Which climate scenario to anchor to (default: a mid-range scenario such as ~2.5°C with stated-policies transition)
If horizon definitions, materiality threshold, or scenario are missing, apply the defaults above and state each assumption explicitly in the output.
Instructions
Step 1: Classify the Risk in the TCFD Taxonomy
Place the risk in exactly one primary category and note any secondary category:
Physical risks
- Acute: Event-driven — floods, wildfires, cyclones, heatwaves, drought events
- Chronic: Long-term shifts — rising temperatures, sea-level rise, changing precipitation, water stress
Transition risks
- Policy & legal: Carbon pricing, emissions regulation, mandates, litigation
- Technology: Substitution by lower-carbon alternatives, stranded technology
- Market: Shifting demand, input cost volatility, changing customer preferences
- Reputation: Stakeholder sentiment, sector stigmatization, financing access
State the classification and a one-line rationale for why it fits that category.
Step 2: Score Likelihood per Horizon (1–5)
Assess the probability the risk materializes meaningfully in each horizon. Likelihood often rises across horizons for chronic/transition risks and is more horizon-flat for acute physical risks.
| Score | Likelihood | Descriptor |
|---|---|---|
| 1 | Rare | <10% chance in horizon |
| 2 | Unlikely | 10–30% |
| 3 | Possible | 30–55% |
| 4 | Likely | 55–80% |
| 5 | Almost certain | >80% |
Assign a likelihood score for short, medium, and long term, each with a one-sentence justification tied to the scenario assumption.
Step 3: Score Financial Impact per Horizon (1–5)
Estimate gross financial impact (pre-mitigation), expressed against the materiality threshold:
| Score | Impact band | Magnitude (relative to threshold) |
|---|---|---|
| 1 | Negligible | <0.5% revenue / immaterial |
| 2 | Minor | ~0.5–1% revenue |
| 3 | Moderate | ~1–3% revenue |
| 4 | Major | ~3–5% revenue |
| 5 | Severe | >5% revenue or going-concern threat |
Note impact channel(s): direct asset damage, increased operating cost, revenue loss, capex/stranded assets, cost of capital, or compliance penalties. Assign an impact score per horizon.
Step 4: Apply Mitigation Adjustment
Translate gross scores to net scores by accounting for existing controls:
- Strong, funded, tested controls: reduce impact score by up to 2 points
- Partial or planned controls: reduce by up to 1 point
- No effective controls: no reduction Never reduce likelihood through mitigation unless the control prevents occurrence (e.g., relocation eliminates flood exposure). State the gross score, the adjustment, and the net score.
Step 5: Compute the Materiality Rating per Horizon
For each horizon, compute Net Likelihood × Net Impact (max 25) and map to a rating band:
| Composite | Rating |
|---|---|
| 1–4 | Low |
| 5–9 | Moderate |
| 10–14 | High |
| 15–25 | Critical |
Present the three horizon scores and identify the peak horizon (highest composite) — the horizon driving the risk's overall materiality.
Step 6: Assign Overall Rating and Narrative Rationale
Set the overall materiality rating to the highest single-horizon band (a risk that is Critical in the long term is a Critical risk to register), unless the user requests a weighted approach. Then write a narrative rationale covering: how the risk was classified, why likelihood and impact trend the way they do across horizons, the dominant impact channel, the effect of current mitigation, and one recommended next action (e.g., quantify, hedge, disclose, monitor).
Output Format
- Structure: Classification header → scoring table → rating verdict → narrative rationale → recommended action
- Length: 500–800 words
- Format: One TCFD classification line, one combined scoring table (horizon × likelihood × impact × composite × rating), then structured prose
- Sections: TCFD Classification → Assumptions Applied → Scoring Table (3 horizons) → Mitigation Adjustment → Overall Materiality Rating → Narrative Rationale → Recommended Next Action
- Tone: Risk officer — precise, evidence-anchored, conservative on uncertainty, explicit about assumptions; no hedging on the final rating
Quality Checklist
Before delivering output, verify:
- [ ] Risk is classified into exactly one primary TCFD category (physical acute/chronic or transition policy/technology/market/reputation)
- [ ] All three horizons have both a likelihood score and an impact score (6 scores total)
- [ ] Every score has a one-sentence justification — no bare numbers
- [ ] Composite scores are correctly computed (likelihood × impact) and mapped to the correct band
- [ ] Any applied defaults (horizon, threshold, scenario) are stated explicitly
- [ ] Overall rating, peak horizon, and one recommended next action are all present
Notes
- Transition risks typically peak in short-to-medium term (policy moves fast); chronic physical risks typically peak in the long term — flag when a risk's profile contradicts this so the user can sanity-check inputs
- Keep gross and net scores visible; auditors and disclosure reviewers want to see the mitigation logic, not just the final number
- This produces a single-risk assessment designed to drop into a register row — run it repeatedly across risks to build a comparable, consistently scored portfolio
- When inputs are too vague to score impact, ask one targeted question about the largest exposed revenue line or asset before assigning a number rather than guessing
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