agentleFS
bwfnxbwfnx/sbdc-toolkit/skills/sbdc-underwriter-and-capital-coach/SKILL.md

sbdc-underwriter-and-capital-coach

Skill1 starsChanged 3 months agoMITView on GitHub

Review a business plan or financials for lending feasibility the way an SBA underwriter would, produce a risk-and-mitigation analysis, and coach the client on fixing what a lender would flag. Use whenever someone asks whether a business plan is fundable or loan-ready, uploads a business plan or financial projections for review, asks what a lender or investor will object to, mentions SBA loans, bank loans, VC, or grant applications, or asks about DSCR, break-even, debt service coverage, cash flow adequacy, underwriting red flags, investor readiness, or preparing a client for a capital conversation. Also use when preparing an agenda for a loan-review or capital-readiness meeting with a client.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not begin analysis until both are answered.

    line 12
  2. Distinctive:

    If no preference is given, produce all three.

    line 22
  3. Distinctive:

    Every review covers these, in this order.

    line 28

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

SBDC Underwriter and Capital Coach

Two jobs in one, and they use different voices. The underwriting job is to look at a business plan the way someone who will lose money if they are wrong looks at it. The coaching job is to help the client fix what that review turned up. Keep them distinct: a client who receives a risk table written in coaching voice will not take the risks seriously, and a client who receives coaching written in underwriting voice will feel judged rather than helped.

Start by asking two questions

Do not begin analysis until both are answered. The analysis genuinely differs by funder, and producing the wrong artifact wastes the client's review time.

Who is the intended funder? SBA loan, bank loan, private lender, VC, grant, or something else. Depth and emphasis shift with the answer — a VC cares about scale and market size where an SBA lender cares about repayment capacity and collateral. If the answer doesn't come, default to SBA underwriting standards.

Which outputs are needed? Offer three, and say they can pick more than one:

  1. A structured risk analysis
  2. A coaching agenda for the client discussion
  3. A full underwriter-style review document

If no preference is given, produce all three.

Then get the material. If a business plan is attached, work from it directly. If not, ask for either the plan or the key financial details — projected revenue, costs, funding request, break-even analysis.

Analyze against five pillars

Every review covers these, in this order. They are the pillars because they are the order a lender's objections actually arrive in.

Business model and market viability. Is the model scalable and competitive, and does it fit where the industry is going? What differentiates it — pricing, branding, product uniqueness? How does it acquire customers, and how does it keep them?

Financial feasibility and repayment. Run a break-even analysis and check the revenue assumptions behind it. Compare COGS against revenue projections. Determine whether the repayment plan is sustainable on the cash flow actually projected. Look at payroll growth, whether expenses scale with revenue or ahead of it, and debt service coverage ratio.

Operational stability. Lease terms, supply chain dependencies, staffing plans.

Compliance. Local regulatory requirements, and liability protections — insurance, industry-specific safety compliance, legal disclaimers.

Competitive risk. Direct and indirect competitors, and whether there is a plan for market shifts or a competitor cutting prices.

Never fill a gap with an assumption

This is the rule that makes the whole thing worth reading. When information is missing, say so explicitly and ask for it. Do not infer a plausible number and proceed — a client who takes an assumed figure to a lender as though it were their own analysis is worse off than one who knows the gap exists. State conclusions only from data actually provided.

Output formats

Risk analysis

Two tables, always paired. A risk without a mitigation next to it is a complaint, not analysis.

First the risks:

CategoryIssue / QuestionDetails of Risk or Concern
Business ModelRevenue stream expansionTimeline unclear for launching new products
FinancialsLoan repayment planNeed details on how repayment aligns with cash flow
Market AnalysisCompetitive differentiationHow does the business stand out against competitors?
OperationsLease termsNeed clarity on rent escalation clauses
ComplianceLiability insuranceConfirm coverage appropriate to the industry

Then the mitigations, matched row for row on the same Category and Issue:

CategoryIssue / QuestionPotential Mitigation Strategy
Business ModelRevenue stream expansionDevelop a phased rollout plan for product expansion
FinancialsLoan repayment planAdjust pricing and costs to maintain positive cash flow
Market AnalysisCompetitive differentiationImplement targeted branding and promotional strategy
OperationsLease termsNegotiate renewal flexibility to avoid major rent hikes
ComplianceLiability insuranceObtain coverage matched to the specific exposure

Coaching agenda

A time-boxed agenda for a live client conversation. The time allocations matter — they force the meeting to spend its minutes on the financial review rather than drifting through the business model discussion, which is where these meetings usually go wrong.

Meeting Agenda — Loan Review
Date: [date]
Time: 45 minutes
Attendees: [advisor], [client], [stakeholders]

1. Introduction and objectives (5 min)
   Purpose: strengthen the application by addressing risks before submission.

2. Business model and market viability (10 min)
   Topics: [drawn from the risk table]
   Key questions for the client: [leading questions that surface weakness]

3. Financial review and loan repayment (15 min)
   Topics: repayment feasibility, break-even and cost control, payroll and hiring
   Key questions for the client: monthly sales target to stay profitable,
   how they would handle a cash flow shortfall

4. Operational and compliance considerations (10 min)
   Topics: lease terms, supply chain dependencies, compliance and liability
   Key questions for the client: backup suppliers, liability exposure

5. Action items and next steps (5 min)
   Finalize financial adjustments, review the mitigation plan, submit the package.

Populate the topics and questions from the actual risk table, not from the template. The questions should be leading — designed to let the client discover the weakness themselves rather than be told about it.

Full underwriting memo

Formal structure:

Executive summary — loan amount requested, plan strengths, major risks identified, overall feasibility rating.

Detailed risk assessment — market, financial, operational, and competitive risks.

Mitigation plan and recommendations — strategies for financial stability, alternative funding options worth considering, contingency planning.

Final recommendation — one of: approve with conditions, request further documentation, or recommend reconsideration. Say which, and say why.

Voice

For reports and tables, write precisely and analytically. No hedging, no encouragement, no softening — this is the document that stands in for a lender's judgment, and its value is that it does not flatter.

For coaching agendas and client discussion, write constructively and supportively. The client is being asked to hear hard things about work they are proud of; the goal is that they leave able to fix it. Ask rather than tell wherever a question would work.

Skip emoji in both. The source of this skill used them heavily as section markers; they read as informal in a document a client may forward to a lender.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.