deal-qualification-check
Skill1 starsChanged 10 days agoMITView on GitHub
Checks how well one B2B deal is qualified from call notes and email history, marking each key element as confirmed, assumed or unknown, and proposes the next questions. Use whenever the user asks whether a deal is real, why it is stuck, what is missing to move it forward, or to review or qualify an opportunity, even if they never say "qualification".
In this repository7 other files it gives its agents.
- account-research-briefSkill
- call-prep-briefSkill
- cold-outreach-emailSkill
- discovery-call-guideSkill
- post-call-follow-upSkill
- re-engagement-emailSkill
- sales-process-guideSkill
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 15
Ask only for what is missing:
- Distinctive: line 16
Call notes, transcripts or email threads for the deal
- Distinctive: line 21
Read everything provided. Quote or closely paraphrase what the buyer said, as opposed to what the seller said.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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 qualification check
Deals stall on the facts nobody confirmed. This check shows what you actually know about a deal, separated from what you hope is true, and turns each gap into a question for the next conversation.
Inputs
Ask only for what is missing:
- Call notes, transcripts or email threads for the deal
- The company, the contacts and their roles
- The current stage in the user's own words
Steps
- Read everything provided. Quote or closely paraphrase what the buyer said, as opposed to what the seller said.
- For each element below, mark it confirmed (the buyer said it or did it), assumed (the seller believes it, with no buyer statement), or unknown, and attach the evidence.
- Identify the two or three gaps most likely to stop the deal.
- Turn each gap into one open question the user can ask next, in a form a buyer finds easy to answer.
- Recommend one next action and flag anything you could not verify from the notes.
Elements
- Problem: what the buyer says is wrong, in their words.
- Impact: what it costs them, with any number the buyer gave. Never supply a number the buyer didn't.
- Decision maker: who can approve, and whether we have spoken to them.
- Decision process: the steps, people and timeline the buyer described.
- Buying criteria: how they will compare options.
- Internal advocate: someone inside who wants this to happen and has influence.
- Alternatives: other vendors, an in-house build, or doing nothing.
- Next step: the last agreed action, with owner and date.
Format
A table with element, status, and evidence, followed by the top gaps, the questions, and the recommended next action. Keep it to one page. A table fits here because the reader scans for gaps.
Rules, and why
- Only the notes count as evidence. A confident score built on missing information gives false comfort and hides the real risk.
- Don't produce a numeric probability or forecast category. Notes can't support one, and a made-up number gets repeated as fact.
- Treat the seller's optimism as assumed. "They love us" without a buyer statement or action is assumed.
- No budget or pricing questions in the suggested questions unless the buyer raised cost. Frame them around impact and process first.
- Say so when the notes are too thin to judge, and list what to collect instead.
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.

