Purpose
Turn scattered nonprofit information — Form 990, website, annual report — into a rigorous, comparable due diligence scorecard that scores the organization across five funding-critical dimensions and surfaces the specific red flags and green flags a grant decision hinges on. Replaces gut-feel vetting with a repeatable diligence process that produces a defensible fund/hold/pass call.
When to Use
Invoke this skill when the user asks to:
- Vet or evaluate a nonprofit before making a grant or major gift
- Score an organization's financial health, leadership, or program impact
- Assess a nonprofit's theory of change or evidence base
- Compare two or more nonprofits competing for the same funding
- Identify red flags or governance concerns in a charity
- Build a due diligence memo for a board, giving circle, or foundation
Inputs Required
Before starting, collect:
- Organization name: Legal name and any DBA/common name
- Available documents: Which sources are on hand (Form 990, website copy, annual report, audited financials, impact report)
- Grant size and type: Dollar amount and whether general operating, program, or capital
- Funding priority: What outcome or population the funder cares about most
- Risk tolerance: Conservative (established orgs only) / Balanced / High (open to early-stage or higher-risk bets)
- Key financial figures if known: Total revenue, total expenses, net assets, program vs. overhead ratio
- Prior relationship: Any existing knowledge, references, or history with the org
If audited financials are unavailable, work from the most recent Form 990 and flag any conclusion that would change with better financial data.
Instructions
Step 1: Establish the Baseline
Confirm the organization's basics before scoring:
- Legal status (501(c)(3) confirmed, in good standing, not revoked)
- Founding year and organizational stage (startup <3 yrs / growth 3–10 yrs / established 10+ yrs)
- Annual budget size tier (Micro <$250K / Small $250K–$2M / Mid $2M–$10M / Large $10M+)
- Mission statement in one sentence — restate it plainly
Match scoring expectations to stage and budget tier. A 2-year-old org should not be penalized on the same reserve benchmarks as a 30-year-old institution.
Step 2: Score the Five Dimensions (0–5 each)
Score each dimension on a 0–5 scale. For every dimension, state the score, cite the specific evidence behind it, and note what would raise or lower it.
Dimension 1 — Financial Health (weight: 30%) Assess sustainability and stewardship:
- Program expense ratio: % of expenses on programs vs. overhead. 65–85% is healthy; below 65% warrants scrutiny; above 90% may signal underinvestment in infrastructure
- Operating reserves: Months of expenses in unrestricted net assets. 3–6 months is healthy; under 1 month is a red flag
- Revenue concentration: Is more than 40% of revenue from a single source? High dependence on one funder or grant is fragility
- Revenue trend: Growing, flat, or declining over 3 years (from 990 comparison)
- Deficit pattern: Recurring deficits vs. a one-time planned draw-down
- Financial controls: Audited statements present, independent finance committee, no restatements
Dimension 2 — Leadership Quality (weight: 20%) Assess governance and executive strength:
- Board composition: Size (7–15 is functional), independence, relevant expertise, no dominance by a single family or interest
- Executive tenure and stability: Founder-dependent risk vs. institutionalized leadership; recent unexplained turnover is a flag
- Compensation reasonableness: CEO pay relative to budget size and sector norms (990 Part VII)
- Succession readiness: Any evidence of a plan or bench strength
- Governance hygiene: Conflict-of-interest policy, whistleblower policy, document retention policy (990 Part VI checkboxes)
Dimension 3 — Program Evidence (weight: 25%) Assess whether the work actually works:
- Outcome vs. output reporting: Do they measure changed lives (outcomes) or just activities delivered (outputs)?
- Evidence tier: Anecdotal → internal tracking → external evaluation → RCT/independent study. Where do they sit?
- Data specificity: Real numbers with denominators vs. vague claims ("thousands served," "lives changed")
- Cost-effectiveness signal: Cost per outcome if derivable
- Honest reporting: Do they acknowledge what isn't working, or only report wins?
Dimension 4 — Theory of Change Clarity (weight: 15%) Assess strategic coherence:
- Logic chain: Is there a clear line from activities → outputs → outcomes → long-term impact?
- Root-cause alignment: Do the programs address the actual driver of the problem, or just symptoms?
- Focus vs. sprawl: Coherent set of programs vs. mission creep and scattered initiatives
- Assumptions stated: Do they name the conditions their model depends on?
Dimension 5 — DEI Practices (weight: 10%) Assess equity in who leads, who's served, and how:
- Leadership representation: Do board and staff reflect the communities served?
- Proximity to the issue: Are people with lived experience in decision-making roles, not just as beneficiaries?
- Equity in program design: Is the work designed with the community or delivered to it?
- Transparency: Any public commitment, demographic data, or equity practices disclosed?
Step 3: Compile Red Flags and Green Flags
List every flag surfaced during scoring, ranked by severity/strength:
Red flags (rank Critical / Serious / Watch):
- Critical examples: revoked status, going-concern audit note, sub-1-month reserves, executive misconduct, single-funder dependence above 60%
- Serious examples: recurring deficits, overhead below 60%, opaque outcome reporting, board dominated by insiders
- Watch examples: recent leadership turnover, flat revenue, missing succession plan
Green flags (rank Strong / Notable):
- Strong examples: external evaluation showing impact, 6+ months reserves, diversified revenue, community-led governance
- Notable examples: honest failure reporting, clean audits, tenured stable leadership
Step 4: Calculate the Weighted Score and Recommendation
- Multiply each dimension score by its weight, sum to a weighted total out of 5.0, and convert to a 0–100 scale
- Map to a recommendation, adjusted for the funder's stated risk tolerance:
- Fund (typically 3.5+ weighted, no Critical red flags): proceed with the grant
- Fund with conditions (2.8–3.5, or high score with a Serious flag): fund contingent on named safeguards (reporting, restricted use, site visit)
- Hold (2.5–2.8, or one Critical flag that may be resolvable): request specific missing information before deciding
- Pass (below 2.5, or unresolvable Critical flag): decline
- State the recommendation plainly and name the one factor that most drove it.
Step 5: Name the Diligence Gaps
List the 3–5 most important questions still unanswered — the specific documents to request, references to call, or numbers to verify before committing funds.
Output Format
- Structure: Baseline summary → five-dimension scorecard (score + evidence + rationale per dimension) → red flags and green flags (ranked) → weighted score with recommendation → diligence gaps
- Length: 1,000–1,400 words
- Format: Structured document; scorecard as a summary table plus dimension detail; flags as ranked bulleted lists
- Sections: Organization Baseline → Dimension Scorecard (5 dimensions) → Red Flags → Green Flags → Weighted Score & Recommendation → Outstanding Diligence Questions
- Tone: Program officer's diligence memo — evidence-driven, fair, unsentimental; specific enough to defend to a board and honest about uncertainty
Quality Checklist
Before delivering output, verify:
- [ ] All five dimensions are scored 0–5 with specific cited evidence — not adjectives alone
- [ ] The weighted score is calculated correctly using the stated weights (30/20/25/15/10) and shown out of 5.0 and 0–100
- [ ] Every red and green flag is ranked by severity/strength, not listed flat
- [ ] The recommendation is one of the five defined tiers and names its single biggest driver
- [ ] Scoring benchmarks are adjusted for the org's stage and budget tier
- [ ] Outstanding diligence questions name specific documents, numbers, or references to obtain
Notes
- Overhead ratio is the most overused and least reliable single metric — weight financial trend, reserves, and revenue concentration more heavily than program-expense percentage alone
- A missing document is a diligence gap, not automatically a red flag; distinguish "the org is weak here" from "we can't yet tell"
- Founder-led organizations often score high on energy and low on succession readiness — flag key-person risk explicitly rather than letting it hide inside leadership
- When comparing multiple organizations, run each through the full scorecard and rank by weighted score, then let Critical red flags override raw ranking
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