Claude for Winning Proposals: 31 Claude Skills

What these skills are

Winning the work when the client has AI too, as installable Claude skills.

This is Claude for one deal, from the first signal to the signed statement of work — and for the fact that the buyer on the other side has AI too. Not a sales machine: the qualification, the preparation, the notes and the first drafts behind conversations you have yourself, in the order they actually happen.

The skills follow one deal in eight steps, and each step hands a named artifact to the next. Qualifying before you meet hands over MEDDIC marked known and unknown with the gaps turned into call questions, a map of who really decides on their side, and a bid or no-bid call with its reason written down. Reading the brief hands over what the client actually decided and what is generic in a brief written with AI, a pre-call research brief with a source for every fact and the assumptions named as assumptions, an agenda they agreed to with “no fit” among the outcomes, questions by SPIN type that reach cost and who decides, and what staying put costs in their own figures. Turning the call into their problem hands over notes with their words marked verbatim and your inferences kept separate, a recap email that proves you listened before you write anything, a problem statement in their language rather than your firm's, and an issue tree that shows where an answer could be wrong. Options and scope hand over three options from your own price list with no decoy, a fee model chosen for this work, and a MoSCoW scope with the won't list written out. The proposal hands over a short answer-first document, two or three win themes that survive being compared with every other firm, a case study paragraph with approved results only, a plain statement of how you use AI, and a statement of work with acceptance criteria. Then the review: the draft read from the client's chair, a walk-through deck, and one page for the person who signs. Objections start with the AI one — what AI does well here said first — then a like-for-like comparison against a cheaper quote, the objections that come back every deal, and a negotiation plan with a walk-away point. Last, the follow-up: a mutual action plan, two useful follow-ups, a neutral decision brief your contact can take inside, and a win/loss review whichever way it went.

The red line: Claude prepares you to listen better and write faster. It never invents a reference, a result, a number or a client quote, never joins or records a call without everyone's consent, and you send the proposal yourself.

Proposals are mostly won or lost in discovery, so the pack spends its first half there: three of the eight steps, and twelve of the thirty-one skills, come before you write a word of the proposal.

Anything presented as the client's words is marked verbatim and comes from your notes or from a transcript made with everyone's consent. Every fact has a source or is marked as an assumption, and every number, price, result and reference comes from you. Options, never a single price, and never a decoy option built to push another. Opportunities, briefs, options and drafts get assessed here; people are never scored or profiled.

Where a method is a named one — MEDDIC for qualification, SPIN-shaped questions for a discovery call, MoSCoW for scope, an issue tree for the diagnosis — the skill says so and uses it as it was written. Each skill is one artifact or one named method: the inputs it needs, where it breaks, a fixed output template, a finish line and the next skill to run.

Mechanically, each skill is one folder with a SKILL.md file. The 31 are grouped into the eight steps, and each one gives full value on its own: run one, run a step, or work a whole deal through in the order it arrives.

Download all 31 skills. One zip: ready-to-install skill zips, readable SKILL.md files, and the sources behind every method. Free, no signup. Download the zip · Browse on GitHub

The 31 skills you get

The set follows one deal in eight steps: qualify before you meet, read the brief and prepare the call, turn the call into their problem, options and scope, the proposal and the statement of work, review it and walk them through it, objections with the AI one first — then follow up, help them decide and learn from the result.

Where to start, and how the skills chain

Start where the deal actually is. A lead that keeps showing interest and never closes? Start at MEDDIC Qualification, which names what you still do not know and hands you the questions for the next call. A brief that reads complete and cannot be scoped from? AI-Written Brief Review. A call on Thursday? Pre-Call Research Brief, then the agenda and the questions. A call that already happened? Discovery Call Notes, before the words go cold. They said yes? Statement of Work. You lost it? Win/Loss Review. Every skill lists the inputs it needs, so it tells you when the step before it is the one that is actually missing.

They chain. disc-meddic-qualification names the unknowns, disc-buying-committee-map names the people who have to be convinced, and disc-bid-no-bid decides whether to pitch at all. disc-ai-brief-review and disc-pre-call-brief feed disc-discovery-call-agenda and disc-discovery-call-questions, with disc-cost-of-inaction for the deal nobody treats as urgent. disc-discovery-call-notes is the one everything downstream reads: disc-discovery-recap-email proves you listened, disc-problem-statement puts their problem in their own words and disc-issue-tree shows where an answer could be wrong. disc-good-better-best-options, disc-fee-structure and disc-moscow-scope set what is on offer; disc-consulting-proposal writes it, with disc-win-themes, disc-case-study-paragraph and disc-ai-use-statement supplying the parts a model cannot generate, and disc-statement-of-work holding the scope once they say yes. disc-proposal-red-team reads the draft from the client's chair before disc-proposal-walkthrough-deck takes it to the room and disc-executive-summary to the person who signs. disc-ai-objection-answer, disc-quote-comparison, disc-objection-handling and disc-negotiation-plan are for the conversation after. Then disc-mutual-action-plan, disc-proposal-follow-up and disc-client-decision-brief get you a decision, and disc-win-loss-review turns it into the next pitch.

The sources behind each method are in resources/evidence-and-sources.md, inside the pack.

Setup guide

1. Download the pack. One zip: an install folder with 31 ready-to-upload skill zips, a skills folder with the same 31 skills as readable SKILL.md files, and resources/evidence-and-sources.md, which names the source behind each method.
2. Install your skills. In Claude Code, add the marketplace and install winning-proposals-pack, and all 31 load at once. In Claude, turn on code execution in Settings, then Capabilities, then go to Customize, then Skills and upload one zip per skill from the install folder. Prefer working from files? Add the SKILL.md files to your Project knowledge instead; it works, just less cleanly. Keep client names, your price list and anything commercially sensitive out of Claude memory.
3. Bring one deal you are pitching now, and start where it actually is. Pick a live opportunity and run MEDDIC Qualification on it first: it names what you still do not know and turns the gaps into questions for the next call. Keep that deal in one Claude Project with the brief, your call notes and your own price list, so the proposal is written from the same facts as the recap email. Then follow the chain — every skill ends by naming the next one to run.

Where to start

The quality bar

Every skill in the pack holds the same standard:

• Their words stay theirs: anything presented as the client's words is marked verbatim and comes from your notes or a transcript made with everyone's consent
• Every fact has a source or is marked as an assumption; every number, price, result and reference comes from you
• Options, not one price, and never a decoy option built to push another
• The AI question is answered honestly: say what AI does well here, never sell against it, never hide your own use
• Opportunities, briefs, options and drafts are assessed; people are never scored or profiled
• A named person decides, and you send everything
• The red line: Claude prepares you to listen better and write faster; it never invents a reference, a result, a number or a client quote, never joins or records a call without everyone's consent, and you send the proposal yourself

Who made this

Polar Bear is a people ops consultancy for human-size teams (20 to 200 people). Built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. This pack is the free, self-serve version of how we work.

This pack is for the person who pitches the work and then delivers it. If you run a consultancy, a boutique or an agency and want your team working this way — AI carrying the qualification, the preparation and the first draft so people do the listening and the deciding — that's what we build with clients.

Meet Pauline. Want your team winning work this way? Bring the deal you are stuck on and we will pick the skills for it. Book a 30-minute call · Pauline on LinkedIn

Install in Claude Code

Two lines, and every skill in the pack loads at once.

/plugin marketplace add polar-bear-org/claude-skills
/plugin install winning-proposals-pack@polar-bear-skills

Using Claude on the web instead? Download the zip and upload each skill from its install folder under Customize → Skills.