M And A Diligence Checker

M&A Diligence Checker

Purpose

Turn the deal structure and risk allocation into a focused diligence process that distinguishes verified fact, document gap, legal issue, commercial concern, and recommended transaction response.

Required inputs

Obtain:

• deal type, stage, structure, value, jurisdictions, parties, and user's side;
• target group chart, business model, key assets, regulated activities, and material locations;
• agreed scope, materiality, red-flag threshold, lookback period, exclusions, and deadline;
• term sheet, draft transaction documents, prior reports, data-room index, Q&A, and disclosure materials; and
• known sensitivities such as founder dependence, IP ownership, licences, customer concentration, debt, litigation, data, employment, real estate, or tax.

Treat side, structure, scope, materiality, and available data-room universe as blocking. If scope is not agreed, propose a tiered scope and obtain confirmation before calling the review complete.

Method

1. Translate the transaction into diligence hypotheses. Identify what must be true for title, control, value, operation, financing, integration, and planned exit to work.
2. Create a tailored request list rather than a generic dump. Cover only relevant domains: corporate records and capitalisation; ownership and title; financing and security; material contracts; regulatory; litigation; employment and benefits; IP and technology; privacy and cybersecurity; real estate; insurance; tax; environment; anti-bribery, sanctions and other compliance.
3. Maintain a document inventory with request number, period, entity, status, version, response, reviewer, and follow-up. Distinguish Not provided, Not applicable, Provided but incomplete, and Reviewed.
4. Verify corporate existence, authority, ownership, securities, options, convertibles, liens, transfer restrictions, minority rights, and discrepancies between registers, agreements, filings, and the cap table.
5. Review each material relationship for term, economics, change of control, assignment, termination, exclusivity, liability, indemnity, non-compete, consent, breach, dispute, and dependency on a person or asset.
6. Link every issue to evidence and impact. Record entity, document, clause or source, fact, legal dependency, severity, likelihood, value or operational effect, owner, and follow-up.
7. Convert findings into deal responses: condition precedent, consent, pre-closing covenant, price adjustment, escrow or holdback, specific indemnity, warranty, disclosure, remediation, integration plan, or decision not to proceed.
8. Test management explanations against documents and public or official records where authorised. Mark oral explanations as unverified until supported.
9. Update the issue log as documents arrive. Close an issue only with evidence and preserve the audit trail of why its status changed.
10. State review limitations clearly: unavailable documents, sampling, jurisdictions not covered, specialist advice required, reliance, and cut-off date.

Output

Produce:

1. Diligence scope and request list.
2. Data-room completeness tracker.
3. Issue log — severity, evidence, impact, recommendation, owner, and status.
4. Executive red-flag report, prioritised by deal impact rather than document order.
5. Transaction-document matrix mapping findings to conditions, covenants, warranties, indemnities, disclosure, price, and closing deliverables.

Guardrails

• Do not describe the target as clean or compliant beyond the agreed scope and evidence reviewed.
• Do not treat a missing document as proof of a breach, or a management assurance as documentary verification.
• Do not invent public-record checks, licences, ownership, litigation status, or financial consequences.
• Do not disclose privileged, competitively sensitive, personal, clean-team, or restricted data outside authorised review channels.
• Do not give tax, accounting, technical, environmental, or foreign-law conclusions without the appropriate specialist source or verification.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: rohasnagpal
• Source: rohasnagpal/legal-ai-skills
• License: MIT

Install and usage instructions live in the source repository linked above.