Claude CoWork for ESG Sustainability Analysts

What is Claude CoWork?

Claude CoWork is the practice of using Claude as a persistent, framework-aware co-worker embedded in your ESG analysis workflow. This is not about asking a general AI to summarize a sustainability report. It is about configuring Claude with your framework alignment, materiality standards, and disclosure requirements so it produces structured ESG output — KPI extractions mapped to GRI or SASB, risk scores tied to materiality matrices, disclosure drafts that reflect CSRD and TCFD pillars — that you can actually work with.

Claude-native prompts. The prompts in this guide use Claude's native XML tag structure (<context>, <instructions>, <format>, <avoid>) for more precise, consistent output. These tags help Claude parse your intent with less ambiguity. They work in ChatGPT too, but are optimized for Claude.

ESG disclosure is no longer voluntary. The SEC's climate risk rules, EU CSRD (mandatory for large EU-listed companies from FY2024), and UK SECR have made structured ESG reporting a compliance function, not a communications function. Specialized tools like Briink and C3 AI ESG exist, but most analysts are still combining generic AI with Excel and manually reconciling framework mappings. Claude, configured correctly, bridges that gap for extraction, scoring, and first-draft disclosure — with the analyst's judgment driving the outputs that matter.

This guide walks you through configuring Claude for ESG analysis, the five workflows that compress the most time, and the anti-greenwashing guardrails that protect the integrity of every output you sign off on.

Install the ESG Sustainability Analyst Plugin

This guide works on three Claude surfaces. The plugin is the fastest path on two of them. Pick whichever you use:

If you're on Cowork (desktop or mobile app)

Claude Cowork is Anthropic's agentic workspace — Claude completes work autonomously and returns finished deliverables. The ESG Sustainability Analyst plugin packages the workflows below as native skills and slash commands.

1. Open the Cowork plugin directory in your desktop app.
2. Filter by Cowork, search for "ESG Sustainability Analyst", and click Install.
3. The plugin's slash commands and ambient skills are now available in any Cowork task.

If you don't see the plugin in the directory yet, install via custom marketplace: paste https://github.com/alexclowe/awesome-claude-cowork-plugins in your Cowork plugin settings.

If you're on Claude Code (CLI)

Install from your terminal:

claude plugin add alexclowe/awesome-claude-cowork-plugins/esg-sustainability-analyst

The plugin's slash commands and skills load on next session.

If you're on Claude.ai (web chat only)

Plugins aren't directly installable on the web chat surface. You have two options:

1. Use the prompts in this guide directly in a Claude Project (covered in the next section). Same outputs, more typing.
2. Upload the plugin's skills as a zip via Settings → Features → Custom Skills (Pro/Max/Team/Enterprise plans). Higher friction; only worth it if you want the auto-activating skills, not the slash commands.

What the plugin gives you (any surface)

Slash command	What it does
/extract-esg-data	Ingest a corporate report and extract ESG KPIs (carbon, diversity, governance) mapped to GRI/SASB/TCFD/CSRD
/assess-esg-risk	Map a company against ESG risk frameworks, flag material issues per SASB materiality map, and generate a risk score
/draft-esg-disclosure	Generate SEC Rule 10b5-1 disclosures, ESG proxy filings, and sustainability report drafts in CSRD format
/track-esg-performance	Ingest historical data, benchmark against a peer set, and track progress toward stated targets

Auto-activating skills (no command needed — Claude applies them when relevant):

• Regulatory Standard Mapping — Auto-map ESG initiatives to GRI/SASB/TCFD/CSRD/EU Taxonomy/SFDR requirements
• Carbon Accounting Automation — Ingest Scope 1/2/3 inputs, calculate footprint per GHG Protocol, and flag data-quality issues

The plugin works standalone for one-off tasks. Pair it with the surface-specific setup below for persistent context across every task — that combination is the full Claude CoWork setup.

Setting Up Claude for ESG Sustainability Analyst Work

Surface note: The Project setup below is for claude.ai web users. Cowork users have their own task-context mechanism (set context once when starting a Cowork task). Claude Code users get the plugin's ambient skills automatically — no Project setup needed. The workflows themselves are surface-agnostic — paste the prompts wherever you're working. The key to consistent, framework-grounded output is using Claude Projects. A Project stores your framework preferences and materiality standards across every conversation.

Step 1: Create an ESG Analysis Project. In Claude, click "Projects" and create one called "ESG Analysis — [Team or Portfolio Name]."

Step 2: Set your custom instructions. In the Project settings, add:

You are my ESG analysis and disclosure assistant. Here is my context:

<analyst-profile>
- Role: [ESG Analyst / Sustainability Manager / Portfolio Analyst / Consultant]
- Organization type: [Corporate sustainability team / Asset manager / Consultancy]
- Primary frameworks: [GRI / SASB / TCFD / CSRD / IFRS S1-S2 / GHG Protocol]
- Reporting jurisdiction(s): [EU / US SEC / UK / Global]
- Sector focus: [List primary sectors — e.g., Consumer Staples, Financials, Energy]
- Materiality approach: [SASB-based / Double materiality (ESRS) / Single materiality]
</analyst-profile>

<rules>
- All ESG outputs are analytical drafts requiring expert review and verification against primary source data before use in any disclosure or investment decision
- Never present calculated figures (emissions, scores, percentages) as verified without noting the data quality tier of the underlying inputs
- All disclosure language must comply with applicable frameworks; flag any claim that cannot be substantiated by the data provided as a greenwashing risk
- Do not substitute Claude's output for legal, audit, or assurance review — flag where independent verification is required
</rules>

Step 3: Upload your materiality matrix and framework crosswalks. Add your SASB materiality map for your primary sectors, your TCFD alignment template, and any internal ESG scoring rubrics to the Project knowledge base.

Step 4: Load GHG Protocol Scope 2 guidance and AR6 GWPs. Paste or upload your organization's preferred emission factor sources and the IPCC AR6 Global Warming Potentials so Claude references the right conversion values.

Step 5: Always work inside this Project. Every new conversation inherits your framework context automatically.

Five High-Leverage Workflows

1. ESG KPI Extraction and Framework Mapping

Extracting structured KPIs from a corporate sustainability report and mapping them to disclosure frameworks is the most time-intensive step in any ESG due diligence or benchmarking exercise. Claude does the initial extraction and gap identification.

<context>
I am pasting text from pages 34-52 of [Company Name]'s FY2024 Sustainability Report covering environmental performance. The company operates in the food retail sector (SASB industry: FB-FR). I need to map reported metrics against GRI 305-1 (Scope 1 emissions), GRI 305-2 (Scope 2 emissions), SASB FB-FR-110a.1 (energy consumed), TCFD Metrics & Targets pillar, and CSRD ESRS E1 (climate change).
</context>

<instructions>
Extract all quantitative and qualitative ESG disclosures from the pasted text. For each:
1. Identify the metric name, reported value, unit, and reporting year
2. Map to the relevant GRI indicator, SASB code, TCFD pillar, or ESRS datapoint
3. Flag where a reported metric partially meets a framework requirement (partial disclosure)
4. Flag where a required framework metric has no reported data (gap)
5. Note any emissions figures that lack a stated calculation methodology or emission factor source — these are data-quality risks
</instructions>

<format>
Table with columns: Metric | Reported Value | Unit | GRI Ref | SASB Code | TCFD Pillar | ESRS Datapoint | Status (Full / Partial / Gap) | Data Quality Notes. Summary paragraph of gaps after the table. Under 600 words excluding the table.
</format>

<avoid>
Inferring values not stated in the source text; treating partial disclosures as compliant; mapping to framework requirements without citing the specific indicator code.
</avoid>

Before Claude: 4-6 hours extracting KPIs and manually cross-referencing four framework mapping tables. After Claude: 20 minutes to paste the relevant report sections, 30 minutes to review the extraction and validate against source pages.

2. ESG Risk Score Against SASB Materiality

Scoring a target company's ESG risk profile against SASB's industry-specific materiality map provides a structured basis for investment screening or supplier assessment. Claude builds the scoring framework and populates it from the inputs you provide.

<context>
Target company: [Company Name], mid-cap specialty chemicals manufacturer. SASB industry: EM-CM (Chemicals). Available data: FY2024 sustainability report (pasted below), CDP Climate disclosure (Score B-), MSCI ESG rating (BB). I am assessing against SASB EM-CM material topics: GHG Emissions (EM-CM-110a), Air Quality (EM-CM-120a), Hazardous Waste (EM-CM-150a), Employee Health & Safety (EM-CM-320a), Product Design for Use-Phase Efficiency (EM-CM-410a).
</context>

<instructions>
For each SASB material topic:
1. State the specific SASB disclosure metric(s) for EM-CM
2. Extract what the company has reported against each metric
3. Identify the disclosure gap or data quality issue
4. Assign a risk tier: Low (strong disclosure + performance), Medium (partial disclosure or lagging performance), High (no disclosure or significant performance concern)
5. Provide a brief justification tied to the specific SASB indicator
Produce an aggregate ESG risk score (weighted average) with your weighting rationale.
</instructions>

<format>
Scoring table: SASB Topic | Metric Code | Reported Data | Gap / Issue | Risk Tier (L/M/H). Weighting rationale as a short paragraph. Aggregate score as a number (1=Low, 3=High) with confidence note. Total under 500 words excluding table.
</format>

<avoid>
Assigning a risk tier without citing the specific SASB indicator; treating absence of disclosure as equivalent to good performance; presenting the aggregate score as investment advice or a buy/sell signal.
</avoid>

Before Claude: 3-4 hours manually populating a SASB materiality scorecard from multiple sources. After Claude: 15 minutes to assemble inputs, 30 minutes to review the scoring and apply analyst judgment on the risk tiers.

3. SEC / CSRD-Format Disclosure Draft

First-draft disclosure language reduces the blank-page problem for regulated filings. Claude generates structured draft language aligned to the specific rule and section you specify, with required elements and placeholder flags.

<context>
I am drafting the climate risk section for a US-listed company's 10-K, aligned to SEC climate risk disclosure requirements and informed by TCFD. The company is a logistics firm with material Scope 1 emissions (fleet) and significant Scope 3 Category 4 (upstream transportation). They have set a 2030 net-zero target aligned to SBTi 1.5°C pathway. I am also preparing a parallel CSRD ESRS E1 disclosure for their EU subsidiary.
</context>

<instructions>
Draft two disclosure sections:
1. SEC 10-K climate risk section: Governance (board oversight, management role), Strategy (climate-related risks and opportunities with short/medium/long-term horizon), Risk Management (identification and assessment process), Metrics & Targets (Scope 1, 2, 3 with GHG Protocol Scope 2 dual reporting — market-based and location-based, and SBTi target statement)
2. CSRD ESRS E1 section: Transition plan (SBTi alignment), Physical risk assessment (acute and chronic), Scope 1/2/3 per GHG Protocol including Scope 2 dual reporting requirement under ESRS E1-6, carbon removal strategy note
Flag every placeholder where company-specific data must be inserted before filing.
</instructions>

<format>
Each disclosure as a separate section with H3 headers matching the framework pillar or ESRS datapoint label. Placeholders in [BRACKETS]. Anti-greenwashing flags in bold: note any claim in the draft that requires third-party assurance or cannot be made without supporting data. Under 900 words.
</format>

<avoid>
Asserting "net zero" or "carbon neutral" claims without scope definition and timeline; omitting Scope 3 material categories identified in the context; drafting language that implies regulatory compliance without noting that legal review is required.
</avoid>

Before Claude: 6-8 hours drafting a first pass across TCFD pillars and ESRS E1 datapoints. After Claude: 20 minutes to brief the company context, 60 minutes to review, insert verified data, and flag for legal review.

4. Scope 1/2/3 Carbon Footprint with Data-Quality Flags

Carbon accounting requires source-data traceability and explicit data-quality tiers. Claude structures the calculation from your activity data, applies the correct GWPs, and flags where estimates introduce material uncertainty.

<context>
Company: mid-size manufacturing firm, 2 facilities (EU + US). Available data:
- Scope 1: Natural gas consumption 12,500 MWh (facility metered), diesel fleet 180,000 litres (fuel card records), HFC-134a refrigerant loss 45 kg (engineer estimate)
- Scope 2: Purchased electricity 8,200 MWh EU facility (supplier invoice), 6,100 MWh US facility (utility invoice). EU supplier carbon intensity: 82 gCO2e/kWh (contractual). EU grid average: 295 gCO2e/kWh. US grid average (eGRID region): 410 gCO2e/kWh.
- Scope 3: Business travel (flights, actuals from travel management company), employee commute (survey-based estimate, 320 employees), purchased goods (spend-based, $4.2M direct materials spend).
Use IPCC AR6 GWPs (CH4 = 29.8, HFC-134a = 1526). Apply GHG Protocol Scope 2 dual reporting.
</context>

<instructions>
Calculate Scope 1, Scope 2 (market-based and location-based), and Scope 3 (Categories 6, 7, 1) in tCO2e. For each:
1. Show the activity data × emission factor × GWP calculation
2. State the emission factor source
3. Assign a data-quality tier: Primary (metered/invoiced), Secondary (survey/estimate), Spend-based (highest uncertainty)
Flag where spend-based or survey estimates contribute >10% of total inventory — these require improvement in future reporting cycles per GHG Protocol guidance.
</instructions>

<format>
Calculation table: Scope | Category | Activity Data | Emission Factor | GWP | tCO2e | Data Quality Tier | Source. Scope 2 dual reporting as two rows (market-based, location-based). Total inventory summary. Data-quality flag summary as a short bulleted list. Under 400 words excluding tables.
</format>

<avoid>
Using AR5 GWPs when AR6 has been specified; conflating market-based and location-based Scope 2 in a single figure; omitting data-quality tiers; presenting spend-based Scope 3 estimates without the inherent uncertainty caveat.
</avoid>

Before Claude: 4-5 hours building a carbon accounting model from scratch in Excel with manual GWP lookups. After Claude: 15 minutes to input activity data, 45 minutes to verify emission factors against primary sources and review the calculation structure.

5. ESG Portfolio Benchmarking

Comparing a portfolio's ESG performance against a peer set requires a consistent scoring basis and a clear view of where companies diverge. Claude structures the peer comparison from your input data and identifies the gaps that matter for engagement or exclusion decisions.

<context>
I am benchmarking 5 consumer staples companies in our portfolio against each other on climate and labor metrics. SASB industry: FB-FR (Food Retail) for 3 companies, CN-FB (Food & Beverage) for 2. Metrics: Scope 1+2 emissions intensity (tCO2e/revenue), renewable energy %, employee injury rate (TRIR), supplier audit coverage %, board ESG oversight disclosure. Data pasted below from each company's most recent sustainability report or CDP disclosure. Two companies have IFRS S2-aligned disclosures; three do not.
</context>

<instructions>
For each company:
1. Extract the stated value for each metric; note year reported
2. Flag where the metric is not reported or uses a non-comparable methodology
3. Identify the two companies with strongest and weakest performance per metric
4. Note where IFRS S2 alignment creates a comparability advantage
5. Produce an engagement priority ranking: which 2 companies have the largest performance gap on material metrics relative to peers, and what specific disclosure improvement should we request?
</instructions>

<format>
Benchmarking table: Company | Scope 1+2 Intensity | Renewable % | TRIR | Supplier Audit % | Board ESG Oversight | Comparability Notes. Rankings as a short paragraph per metric. Engagement priority list with specific ask per company. Under 500 words excluding table.
</format>

<avoid>
Comparing metrics across companies that use different calculation methodologies without flagging the comparability limitation; treating non-disclosure as equivalent to poor performance; framing engagement recommendations as divestment advice.
</avoid>

Before Claude: 5-6 hours building a peer benchmarking table and writing the engagement memo. After Claude: 20 minutes to paste company data, 40 minutes to review the benchmarking output and draft the engagement letter.

What This Looks Like in Your Week

Monday: Kick off any new due diligence by running Workflow 1 on the target company's latest sustainability report. Extract and map KPIs before you do anything else — gaps you find now shape the questions you ask later.

Tuesday/Wednesday: Run the risk score (Workflow 2) and carbon accounting (Workflow 4) using the KPIs extracted Monday. Keep your data-quality tiers documented — they become your materiality caveat in the final output.

Thursday: Draft disclosure sections (Workflow 3) using verified data from the week's analysis. Mark every placeholder. Route to legal before Friday.

Friday: If you are running a quarterly portfolio review, batch the peer benchmarking (Workflow 5) for all active positions. Produce engagement priority rankings before the investment committee cycle.

Ongoing: Update your Project instructions when new ESRS datapoints, SASB standard revisions, or updated AR GWPs are published. ESG frameworks change faster than most disclosure teams update their templates.

What to Avoid

Asserting "carbon neutral" or "net zero" without scope and timeline. Claims without GHG Protocol-defined scope boundaries and a credible interim pathway violate the FTC Green Guides and are a primary target of the EU Green Claims Directive. Every disclosure draft Claude produces should flag these as requiring substantiation before filing.

Using AR5 GWPs when AR6 is the current standard. IPCC AR6 updated GWPs significantly (HFC-134a went from 1430 to 1526; CH4 from 25 to 29.8). Verify your Project instructions reference AR6 and update emission factor sources accordingly.

Treating spend-based Scope 3 estimates as primary data. Spend-based Category 1/2 estimates introduce 40-60% uncertainty per GHG Protocol guidance. Flag them as requiring improvement and do not use them as the basis for net-zero target claims without disclosing the data quality limitation.

Omitting Scope 2 dual reporting. ESRS E1-6 and IFRS S2 both require market-based and location-based Scope 2. A disclosure that presents only one number is non-compliant and cannot be corrected by Claude — it requires the right source data upstream.

Skipping assurance review on regulated disclosures. Claude drafts; it does not provide the limited or reasonable assurance that CSRD requires for sustainability statements. Every regulated disclosure that Claude touches needs a documented human review and, where required, third-party assurance before filing.

Resources

• Explore the ESG Sustainability Analyst AI plugin for ready-to-use prompt workflows mapped to GRI, SASB, TCFD, and CSRD
• Browse the ESG sustainability analyst profession guide for a fuller picture of where AI fits across the disclosure lifecycle
• Run the AI readiness audit for ESG analysts to identify your highest-leverage starting point