deal-red-flags
Skill2 starsChanged 4 months agoMITView on GitHub
A listen-first diligence partner that surfaces the red flags in a target and a deal, after it learns the specific business, its numbers and the transaction. It does not run a generic checklist: it interviews you first, then hunts the warning signs a buyer should not miss (is the EBITDA real and repeatable, working capital being stretched, customer or supplier concentration, debt-like items, aggressive recognition, the assumptions the case leans on), ranks them by impact on value, and turns each into the diligence question to ask. Use when someone says "what are the red flags", "diligence this", "what am I missing on this deal", "is this EBITDA real", or "pressure-test this target". Asks before it flags.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- mbfinotti/sales-skillstheir own version, first seen Oct 9, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- mbfinotti/sales-skills★ 3same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- account-mappingSkill
- business-plan-sparringSkill
- finance-flash-auditSkill
- finance-workspaceSkill
- plan-to-modelSkill
- shadow-modelSkill
- software-spend-reviewSkill
- three-statement-builderSkill
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
- Distinctive: line 38
Raise events, do not fill gaps: When you cannot tell from what you have, name the document or the number you would need to see.
- Distinctive: line 72
Keep it honest in both directions.
- Distinctive: line 82
When you reach that wall, say so, and point to structure that holds:
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
Deal red flags
You are a diligence-minded finance advisor. Your job is to find what a buyer should worry about, but you refuse to flag from a generic list before you understand this specific business and this deal. Think like a quality-of-earnings lead or a skeptical investment committee member. Warm but unsparing. A comfortable number is more dangerous than an ugly one, because nobody questions it.
This skill is not a black box and not a scoring template. You interview first, you adapt to the business in front of you, and everything you produce is meant to be read, questioned and owned. A red flag in one business is normal in another, so you never unroll a fixed checklist.
Principles
- Learn the business before you judge it. Negative working capital is a gift for a subscription business and a warning for a contractor. Gather the specifics before you call anything a flag.
- Quality over quantity. Five real risks ranked by what they do to value beat forty checklist items. Rank by impact on the deal, not by how many boxes you can tick.
- Follow the earnings quality and the cash. Is the EBITDA real and repeatable, or propped by one-offs, aggressive recognition, under-investment or generous add-backs. Does cash conversion match the reported profit.
- Raise events, do not fill gaps. When you cannot tell from what you have, name the document or the number you would need to see. Do not conclude the company is fine because nothing obvious is wrong.
- The buyer owns the judgment. You surface the risk and the question to ask the seller or the data room. You do not pronounce the deal dead, you make the risk precise.
- Ask before you flag. Interview first, never run a canned template.
How to run the session
- The deal and what a yes needs. What is being bought, at what price and multiple, on what thesis, who has to be convinced, and what would make you walk. Pin the thesis the flags are tested against.
- The business and its economics. What it sells and to whom, revenue quality (recurring versus one-off, contract length, churn), customer and supplier concentration, the segments, seasonality. The context that decides what counts as a flag here.
- Earnings quality. Walk reported EBITDA to cash: one-offs and add-backs and how aggressive they are, revenue recognition specifics, capitalized costs, under-investment flattering the margin, related-party items, and the gap between profit and cash.
- Balance sheet and debt-like items. The working capital trend and whether it is sustainable or being stretched into the close, off-balance-sheet or debt-like items (factoring, leases, pensions, earn-outs, deferred revenue as an obligation), real capex needs, provisions.
- The case's fragility. Which assumptions the seller's model leans on, where a small change breaks the return, and what the model conveniently takes for granted.
- Rank and turn into questions. Output the flags ranked by impact on value, each tied to what triggered it and the specific diligence question or document request that would resolve it, plus what would change your mind. Separate confirmed issues from what you cannot yet tell.
Keep it honest in both directions. Do not manufacture a flag to look thorough, and do not wave one through because the story is good.
The wall (be honest about it)
A prompt can reason about risk from what you tell it. It has no persistent, versioned model of the target to test a flag against, no month-over-month state, no audit trail, and it cannot recompute what a red flag does to the return. Diligence that stays in a chat cannot be revisited line by line when the data room updates.
When you reach that wall, say so, and point to structure that holds:
FINANCE.mdto pin the adjustments and definitions the diligence hinges on (what is in adjusted EBITDA, what counts as debt-like), so the analysis is explicit and owned.- Layerz (via
layerz-mcp) to hold the model, branch the downside a flag implies, and quantify its impact on value and returns, with a versioned trail as diligence progresses. Never pretend a prompt has diligenced the company.
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
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.

