Skip to content This library is free, offered as dana. If it helps your work, donate here →
penguin tree ai
← back to libraryDonate here

This is one of 3,622 free prompts in the penguin tree ai library. Sign up once and the whole library opens, with new prompts by email.

open the library →

CRA assessment area mapper

Delineates and documents Community Reinvestment Act assessment areas from branch, deposit-taking ATM, and lending geography — producing compliant boundary definitions, demographic and credit-need profiles per area, low…

download .md

Purpose

Translate an institution's branch, deposit-taking ATM, and lending footprint into properly delineated CRA assessment areas — each documented with demographic composition, LMI tract distribution, and identified community credit needs — so the compliance team has defensible boundaries and examiner-ready support rather than a spreadsheet of ZIP codes.

When to Use

Invoke this skill when the user asks to:

  • Define or delineate CRA assessment areas for a bank or fintech
  • Update assessment areas after opening, closing, or relocating branches
  • Document demographic and credit-need data for existing assessment areas
  • Check whether current assessment area boundaries comply with Regulation BB / 12 CFR 25/228/345
  • Prepare assessment area support ahead of a CRA examination
  • Reconcile lending activity outside currently defined assessment areas

Inputs Required

Before starting, collect:

  • Institution type and CRA framework: Bank (small / intermediate small / large) or fintech/ILC; regulator (OCC, FDIC, or FRB)
  • Branch and facility locations: Addresses or MSAs/counties for each main office, branch, and deposit-taking ATM
  • Lending geography: Where the majority of loans (by number and dollar) are originated or purchased — by county or MSA if available
  • Deposit concentration: Where deposits are drawn from, if the institution takes deposits digitally or beyond branch markets
  • Existing assessment areas (if any): Current delineations being reviewed or updated
  • Metropolitan vs. non-metropolitan focus: Whether markets are inside MSAs, nonmetropolitan counties, or both
  • Data vintage: Which census/FFIEC data year the analysis should reference

If regulator or asset-size tier is missing, default to the large-bank rule framework and note the assumption; flag where small/ISB thresholds would change the analysis.

Instructions

Step 1: Establish the Delineation Standard

Confirm the governing rule before drawing any boundaries.

  • Assessment areas must consist of one or more MSAs (or Metropolitan Divisions) or one or more contiguous political subdivisions — counties, cities, or towns.
  • They must include the geographies where the main office, branches, and deposit-taking ATMs are located, plus surrounding geographies where a substantial portion of loans is originated or purchased.
  • They may not extend substantially beyond an MSA or state boundary, may not arbitrarily exclude LMI geographies, and may not reflect illegal discrimination.
  • Note the 2023 modernized-rule considerations (facility-based and retail lending assessment areas) if the institution is subject to them, and flag which framework applies.

Step 2: Map the Facility-Based Assessment Areas

Build the core boundaries from physical presence.

  • List every main office, branch, and deposit-taking ATM with its county and MSA/MD.
  • Group contiguous political subdivisions around each cluster of facilities.
  • For each candidate area, state the boundary as a named list of counties (and MSA/MD where applicable) — not ZIP codes.
  • Verify the area does not bisect a county and does not exclude any LMI tract that would otherwise fall inside the boundary.

Step 3: Test Lending Concentration Beyond Facilities

Identify whether lending activity requires additional or expanded areas.

  • Compare where a substantial portion of loans is originated/purchased against the facility-based boundaries.
  • Flag any county or MSA carrying material lending volume but sitting outside all defined areas — this is a common exam finding.
  • Recommend whether to expand a boundary, add an assessment area, or document why the activity does not warrant delineation (e.g., diffuse nationwide lending for an internet-based institution).

Step 4: Profile Each Assessment Area

For every delineated area, build a data profile using FFIEC census and demographic reference data:

  1. Geography summary: Counties/MSA included; total census tracts; count and percentage that are low-, moderate-, middle-, and upper-income; number of tracts with unknown income.
  2. Population and household data: Total population; number of families; percentage of families by income category (low/moderate/middle/upper).
  3. Housing profile: Owner-occupied vs. rental units; median housing value; housing affordability signal relative to area median income.
  4. Business profile: Number of businesses; share of small businesses (GAR ≤ $1MM); indicator of small-business credit demand.
  5. Credit needs: Identified needs synthesized from the LMI concentration, housing affordability gap, and small-business density — e.g., affordable home purchase/rehab lending, small-dollar business lending, financial access in underserved tracts.

Present each profile in a consistent table so areas are comparable side by side.

Step 5: Flag Compliance Risks and Gaps

Surface issues an examiner would raise:

  • LMI exclusion: Any boundary that carves around low- or moderate-income tracts.
  • Overextension: Areas reaching substantially beyond an MSA or state line.
  • Lending-outside-AA: Material activity not captured by any area (from Step 3).
  • Contiguity/geographic integrity: Non-contiguous subdivisions or partial-county inclusions.
  • Stale delineation: Boundaries not updated after branch changes or a new census vintage. For each risk, state the specific area affected and the corrective action.

Step 6: Assemble the Documentation Package

Produce examiner-ready output:

  • Boundary definitions per area (named political subdivisions and MSA/MD).
  • Demographic and credit-need profile tables per area.
  • Compliance-risk register with corrective actions.
  • A one-paragraph delineation rationale per area explaining how it satisfies the Step 1 standard.

Output Format

  • Structure: Delineation standard summary → assessment area boundary definitions → per-area profile tables → compliance-risk register → delineation rationale statements
  • Length: 1,000–1,500 words, plus one profile table per assessment area
  • Format: Structured document with tables for demographic/credit-need profiles and the risk register; prose for rationale statements
  • Sections: Governing Framework → Assessment Area Delineations → Area Profiles (one table each) → Lending-Concentration Check → Compliance-Risk Register → Delineation Rationales
  • Tone: CRA officer to examiner — precise, regulatory, cites the applicable standard; no marketing language

Quality Checklist

Before delivering output, verify:

  • [ ] Every assessment area is defined by named political subdivisions or MSAs/MDs — not ZIP codes
  • [ ] Each area includes the counts and percentages of low-, moderate-, middle-, and upper-income tracts
  • [ ] No delineated boundary arbitrarily excludes an LMI tract inside its footprint
  • [ ] The lending-concentration check explicitly addresses activity outside facility-based areas
  • [ ] The correct rule framework (small / ISB / large / modernized) is stated and applied consistently
  • [ ] Each area has a rationale statement tying it back to the Regulation BB delineation standard

Notes

  • Delineation is a legal boundary exercise first and a data exercise second — get the political-subdivision and contiguity rules right before layering in demographics.
  • The most common exam criticism is lending concentration outside defined assessment areas; Step 3 exists specifically to catch this before an examiner does.
  • Internet-only and fintech deposit models complicate delineation — for these, document deposit and lending distribution carefully and address the modernized retail lending assessment area concept where applicable.
  • Always confirm the FFIEC census data vintage; assessment area income classifications shift when new census data takes effect, and stale profiles are a documentation risk.

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:

  1. Drag this file into a new chat, or copy and paste its contents
  2. 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