# 39 Claude Skills for Consultants, From the Client's Brief to the Signed Scope A Claude team of 8 desks for independent consultants: read the brief, run discovery, structure the problem, recommend, price, propose and answer the AI question. - Page: https://lantern.meet-polar-bear.com/skills/consultants-pack - Download: https://lantern.meet-polar-bear.com/skills/consultants-pack.zip - Source: https://github.com/polar-bear-org/claude-skills/tree/main/packs/consultants-pack - Price: free ## Install in Claude Code ``` /plugin marketplace add polar-bear-org/claude-skills /plugin install consultants-pack@polar-bear-skills ``` ## The 39 skills you get The set follows one engagement in eight stages, each a desk: reading the brief and deciding to bid, preparing the call and keeping their words, structuring the problem, options and a recommendation, fees and scope, the proposal and the statement of work, presenting and answering, then following up, deciding and reviewing. ### 1 · Intake desk - **AI-Written Brief Review** — use when: The brief looks finished, probably drafted with AI, and you still cannot scope or price from it. Output: Line-by-line reading, six constraint checks, client questions in priority order. - **MEDDIC Qualification** — use when: Prospects take the call and sound keen, and the proposals that follow still do not close. Output: Known and unknown per letter with sources, fit against your practice brief, call questions. - **Buying Committee Map** — use when: You pitched to one friendly contact and the person who signs has never seen you. Output: Six buying roles, stated needs with sources, unmet roles with one question each. - **Bid/No-Bid Decision** — use when: An RFP or an AI-compiled shortlist arrives and you suspect you are the comparison quote. Output: Criteria before facts, shortlist check, cost of pursuit, decision record, no-bid reply. ### 2 · Discovery desk - **Pre-Call Research Brief** — use when: The call is in twenty minutes and you want to walk in with a view, not a fact sheet. Output: Sourced public facts, likely trigger, two or three hypotheses with test and kill answers. - **Discovery Call Agenda** — use when: First calls wander, the wrong roles attend and nobody agreed how the call should end. Output: Agenda note to send ahead, attendees by role, plain consent line for recording. - **Discovery Call Questions** — use when: Your discovery calls gather facts and never reach what the problem costs or who decides. Output: One-page SPIN guide, open gaps first, decision process block, what to listen for. - **Cost of Inaction** — use when: The client enjoys the conversations and nothing makes the work urgent. Output: Consequences in the client's figures, arithmetic shown, timing pressure, questions for blanks. - **Discovery Call Notes** — use when: You took notes in a rush and the proposal will need the client's exact words. Output: Verbatim words, facts, figures, your inferences, contradictions, open questions, next step. - **Discovery Recap Email** — use when: You want the client to confirm you understood before you write any proposal. Output: "Here is what I heard" in their words, their priorities, open points, dated next step. ### 3 · Problem desk - **Problem Statement** — use when: Your proposals describe your practice instead of their problem, and nobody has said which decision the work serves. Output: Point of view in their words, the decision it serves, success measures, impact, rival test. - **Issue Tree** — use when: The client already holds an AI answer and needs someone to show which parts rest on evidence. Output: One stated cut, branches tested or untested, the client's AI answer placed on the tree. - **Hypothesis Test Plan** — use when: You already believe you know the answer and work without a colleague to argue the other side. Output: Hypotheses that can fail, a rival for each, the telling observation, checks with owners. - **Stakeholder Interview Plan** — use when: The room agreed as soon as the most senior person spoke, and you doubt you heard the facts. Output: Whom to hear by role, neutral brief, same core questions, synthesis of unique information. ### 4 · Strategy desk - **Strategic Options** — use when: The client sees two options and dislikes both, or you see one and it happens to be yours. Output: Three to five distinct ways to solve it, including doing nothing and doing it in house. - **Trade-Off Table** — use when: Someone wants the options compared and you will not make up a score to do it. Output: Criteria in the client's terms, consequences with sources, who bears each cost, no score. - **Outside View Check** — use when: Your timeline or result estimate is mostly a story about why this client is different. Output: Forecast stated precisely, comparable cases and their range, your revised estimate. - **Recommendation Pre-Mortem** — use when: You are about to commit to a recommendation and nobody has tried to break it. Output: Failure defined at a date, ways it fails, early signals, prevention, what stays exposed. - **Sensitivity Analysis** — use when: The client asks what would change your mind and you want a straight answer ready. Output: Load-bearing assumptions varied, the switching point, evidence worth getting, a verdict. - **Recommendation Memo** — use when: The analysis is done and the client needs your position on a single page, with your name on it. Output: Your position first, options rejected, costs accepted, what would change it, review date. ### 5 · Pricing desk - **Good-Better-Best Options** — use when: One price gets you a yes or a no, and too often it is a no. Output: Three priced ways to buy the work, what each leaves out, decoy test on every pair. - **Fee Structure Choice** — use when: The client asks why they should pay for your hours now that AI makes the work quicker. Output: Five fee models compared on four counts, protections needed, your answer on hours. - **MoSCoW Scope** — use when: The AI-drafted brief asked for everything and the fee can cover only part of it. Output: Every item on one line, won't list with reasons, effort split, what moves if budget changes. ### 6 · Proposal desk - **Consulting Proposal** — use when: The client needs something they can read in five minutes and pass to people who missed the call. Output: Answer-first proposal with the decision asked, their problem, three options, proof, AI use. - **Win Themes** — use when: The buyer compared advisers with AI and every one of you now sounds the same. Output: Two or three themes with proof, the AI-shortlist test, where each lands in the proposal. - **Case Study Paragraph** — use when: You need proof no model can generate, and only work you really did will count. Output: 120 to 180 word case, a 40-word version, permission status, results as approved. - **AI Use Statement** — use when: You are unsure whether to tell clients you use AI, and they have begun to ask. Output: Task-by-task AI use, who checks and signs, data handling, what you will not use AI for. - **Statement of Work** — use when: The client has agreed and the scope has to survive the new ideas that arrive next month. Output: Results, deliverables with acceptance criteria, dated timeline, won't list, change control. - **Proposal Red Team Review** — use when: You are about to press send and have no colleague to read it the way the buyer will. Output: Cold read as the absent signer, every claim traced or flagged, edit list by severity. ### 7 · Client room - **Proposal Walk-Through Deck** — use when: Proposals you email cold go quiet, so you want to take the client through this one live. Output: Six to ten slides, full-sentence headlines, options side by side, speaker notes, read version. - **Proposal Executive Summary** — use when: The person who signs has never met you and will read exactly one page. Output: Decision asked and date first, situation, complication, question, options in one line each. - **AI Objection Answer** — use when: A client says AI could do this in seconds, or asks what they pay you for now they have Claude. Output: The question behind the words, a three-column task split, what you say, a shared option. - **Like-for-Like Quote Comparison** — use when: The client slides a quote across the table at a fraction of your fee. Output: Both quotes on the same lines, risk per missing line, gaps in your own, neutral questions. - **Objection Handling Map** — use when: The same objections return on every engagement and you improvise the answer each time. Output: One row per objection, the real concern, an honest response, proof held, lines to avoid. - **Negotiation Plan** — use when: The client is haggling and you are tempted to cut the fee just to end the awkwardness. Output: Your alternative and walk-away, tradables with asks, concession order, the pre-discount question. ### 8 · Follow-through desk - **Proposal Follow-Up Emails** — use when: A proposal sits unanswered and every draft you start begins "just checking in". Output: Two follow-ups each with one useful reason, then a short close-the-loop note. - **Client Decision Brief** — use when: Your contact backs the work and has to persuade people you will never meet. Output: Neutral page for your contact, every route including in house, what would make each right. - **Mutual Action Plan** — use when: The client says yes in the meeting and then nothing happens for six weeks. Output: Every step from yes to day one, owners by role, dates, questions for ownerless steps. - **Win/Loss Review** — use when: You lost a pitch, or won one you cannot explain, and want the next one to go better for it. Output: Plan against what happened, causes sorted, where it was decided, changes as concrete edits. ## Workflows - **Decide whether to bid** (A polished brief or an RFP has landed and nothing is quoted yet.) AI-Written Brief Review → MEDDIC Qualification → Buying Committee Map → Bid/No-Bid Decision - **Run the discovery call** (A first call to prepare, capture and play back in the client's words.) Pre-Call Research Brief → Discovery Call Agenda → Discovery Call Questions → Cost of Inaction → Discovery Call Notes → Discovery Recap Email - **Structure the problem** (The client has confirmed the recap, or already holds an AI answer.) Problem Statement → Issue Tree → Hypothesis Test Plan → Stakeholder Interview Plan - **Reach a recommendation** (You need real options and a position you can defend.) Strategic Options → Trade-Off Table → Outside View Check → Recommendation Pre-Mortem → Sensitivity Analysis → Recommendation Memo - **Price, scope and propose** (The route is chosen and the fee, the scope and the proposal are still to write.) Good-Better-Best Options → Fee Structure Choice → MoSCoW Scope → Consulting Proposal → Statement of Work → Proposal Red Team Review - **Present and answer** (The proposal is ready and the AI question, a cheaper quote or haggling is coming.) Proposal Walk-Through Deck → Proposal Executive Summary → AI Objection Answer → Like-for-Like Quote Comparison → Objection Handling Map → Negotiation Plan - **Follow through** (The proposal is out and the client has gone quiet, or has said yes.) Proposal Follow-Up Emails → Client Decision Brief → Mutual Action Plan → Win/Loss Review ## The quality bar - Their words, with a source: a fact, figure or quote appears only if it traces to your notes, your files or the client's own pages. Anything else is a blank with the question that would fill it - Your numbers only: fees, rates, effort and results are yours or the client's. No benchmark, market rate or typical figure is supplied - Honest about AI: what AI does well is said first. Nothing sells against AI, and your own use of it is never hidden. Doing the work in house with the client's own AI is written as a real option - Roles, never ratings: maps and plans record roles and stated positions. Nothing grades, ranks, scores or profiles a person, and options are compared without a weighted score unless you ask for one and own the weights - A decision at the end: every output closes with what a named person decides, and by when - Advisers for legal points: contract, tax, data and recording questions are flagged “check with a qualified adviser”, never answered as advice. Recordings and transcripts are used only with everyone's consent - The red line: Claude structures, drafts and checks; it never invents a reference, a result, a number or a client quote, and you run the conversation, make the recommendation and send everything yourself ## Questions people ask ### What is in the Claude for Consultants · Team in Claude pack? 39 Claude skills arranged as eight desks that follow one engagement: Intake, Discovery, Problem, Strategy, Pricing, Proposal, the Client room and Follow-through. Each skill produces one artifact and names the skill to run next. ### How do I install these Claude skills? In Claude Code, two lines install all 39 at once. In Claude.ai or Claude Desktop, download the zip, open Settings, Skills, Add, and upload the skills you want. Either way it takes about five minutes. ### Do I need Claude Code to use them? No. The skills work in Claude.ai, Claude Desktop and Claude Code. In the app you upload them yourself; in Claude Code one command installs the whole pack. ### Do the skills work with ChatGPT or other assistants? They are built as Claude Skills, and installing them is a Claude feature. Each one is a plain Markdown file, so you can read it and reuse the method elsewhere, but it will not load and trigger by itself outside Claude. ### Will Claude make the recommendation or send the proposal for me? No. Claude structures, drafts and checks. It never invents a reference, a result, a number or a client quote, and you run the conversation, make the recommendation and send everything yourself. ### Is the pack free? Yes. It is free to use inside your company and in client work, and not for resale. There is no signup: the download and the install command are on this page. ## About Written by Polar Bear (https://meet-polar-bear.com/). Published on Lantern (https://lantern.meet-polar-bear.com/), which turns your notes, documents and call transcripts into social post drafts you review and publish.