cold-start-interview

name cold-start-interview description Run the cold-start interview to learn your commercial contracts practice and write your team practice profile. Use on first use of the plugin, when `~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md` is missing or still contains template placeholders, or when the user says "set up the plugin", "configure commercial contracts", "onboard me", or "let's get started". This is the only skill that should run on a fresh install. argument-hint [--redo to re-run on an already-configured plugin] [--check-integrations to re-probe integrations only] [--side sales|purchasing to re-run only the playbook section for one side] /cold-start-interview Runs the cold-start interview. First run writes ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md ; subsequent runs with --redo re-interview and show a diff before overwriting. Instructions Check current state: Read ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md . If it contains [PLACEHOLDER] or [Your Company Name] , proceed with fresh interview. If populated and --redo not passed, ask: "Looks like you're already set up. Want to re-run the interview? This will overwrite ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md (I'll show you a diff first)." Follow the interview script below. Ask for seed docs: Request 5-10 recent signed agreements (more is better, 20 gives a clearer pattern) and (if it exists) an escalation matrix. Accept file paths, Google Drive links, or [CLM] record IDs. Read the seed docs and extract actual playbook positions. Note deltas between stated positions and what was signed. Migration: If a populated CLAUDE.md (no [PLACEHOLDER] markers) exists at ~/.claude/plugins/cache/claude-for-legal/commercial-legal/*/CLAUDE.md but not at the config path, copy it to the config path and show the user what was migrated. Write ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md (create parent directories as needed) per the structure below. Use the lawyer's own words where possible. Show summary + propose next steps: "Here's what I heard — ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md is written. What did I get wrong?" Offer a test review: "Want to throw a contract at me?" If a [CLM] is connected: offer to bulk-load the renewal register --check-integrations Re-runs the integration availability check (CLM, e-signature, document storage, Slack) and updates ## Available integrations in ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md . Does not re-interview. Use when you connect or disconnect an MCP and want the plugin to notice without rerunning the full setup. When probing: only report ✓ if an MCP tool call actually succeeded. Configured-but-untested connectors should be marked ⚪ with a one-line how-to for confirming. Never report ✓ based on .mcp.json declarations alone — that misleads users into thinking something is wired up when it isn't. --side sales / --side purchasing Re-runs only the playbook section of the interview, calibrated to the specified side, and writes the answers to the matching subsection ( ### Sales-side playbook or ### Purchasing-side playbook ) in ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md . Does NOT re-ask practice setting, role, integrations, team details, or the escalation matrix — those are side-agnostic. Use this when (a) you initially picked "both" at setup and want to build the second side now, or (b) you want to rebuild one side without disturbing the other. Updates the **Active side:** marker in ## Playbook to reflect whichever sides are populated after the run ( sales , purchasing , or both ). Examples /commercial-legal:cold-start-interview /commercial-legal:cold-start-interview --redo /commercial-legal:cold-start-interview --check-integrations /commercial-legal:cold-start-interview --side purchasing Purpose You are meeting this commercial contracts team for the first time. Your job is to learn how they do commercial contracts — not how commercial contracts are done in the abstract — and write what you learn into a living practice profile (the plugin config) that every other skill in this plugin reads before it does anything. The lawyer should leave this conversation feeling like they just onboarded a sharp new paralegal who asked exactly the right questions. They should never see a YAML config file. They should see a document about their team that they can edit in plain English. What "cold start" means Read ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md : Does not exist → start the interview. Contains <!-- SETUP PAUSED AT: --> → greet the user and offer to resume from that section. Contains [PLACEHOLDER] or [Your Company Name] markers but no pause comment → the template was never completed; offer to start fresh or resume from wherever the placeholders begin. Populated (no placeholders, no pause comment) → already configured; skip unless --redo or --side <sales|purchasing> . --side flag: playbook-side-only re-interview If invoked as /commercial-legal:cold-start-interview --side sales or --side purchasing , run only Part 2 (the playbook) calibrated to the specified side, and write the answers to the matching section ( ### Sales-side playbook or ### Purchasing-side playbook ). Do NOT re-ask Part 0 (practice setting, role, integrations), Part 1 (team, volume, mix), or Part 3 (escalation matrix) — those are side-agnostic and already populated. If the other side is already populated, leave it untouched. If neither side is populated yet, the flag still works — it builds the requested side and the other stays as a placeholder pointer until you run --side <other> . Update the **Active side:** marker in ## Playbook : if only one side was built, set it to sales or purchasing ; if both are populated after this run, set it to both . The template structure lives at ${CLAUDE_PLUGIN_ROOT}/CLAUDE.md — use it as the section scaffold. Write the completed practice profile to the config path, creating parent directories as needed. If a CLAUDE.md exists at the old cache path ~/.claude/plugins/cache/claude-for-legal/commercial-legal/*/CLAUDE.md but not at the config path, copy it forward to the config path before proceeding. If the user explicitly asks to re-run setup ("let's redo the interview", "my playbook changed"), run it again and show a diff before overwriting. Check for the shared company profile Look for ~/.claude/plugins/config/claude-for-legal/company-profile.md . If it exists: Read it. Show a one-line confirmation: "You're [name], [practice setting], at [company], [industry], operating in [jurisdictions]. Right? (Or say 'update' to change the shared profile.)" If confirmed, skip the company questions — go straight to the plugin-specific ones. If it doesn't exist: You'll be the first plugin this user set up. After the orientation and fork, ask the company questions and write them to the shared profile (per the template at references/company-profile-template.md in the plugin root), then continue with the plugin-specific questions. Tell the user: "I've saved your company profile — the other legal plugins will read it and skip these questions." The company questions that belong in the shared profile (and should NOT be re-asked if it exists): practice setting, company name, industry, what-you-sell, size, jurisdictions, regulators, risk appetite, escalation names. The plugin-specific questions (playbook positions, review framework, house style, supervision model, etc.) stay per-plugin. Install scope check Before the orientation, if you notice the working directory is inside a project (not the user's home directory), flag it. Say once: Heads up — it looks like this plugin may be project-scoped, which means I can only read files in [current directory]. If you'll want me to read documents from elsewhere (Downloads, Documents, Dropbox), install user-scoped instead — see QUICKSTART.md. You can continue with project scope, but you'll need to move files into this folder. Ask the user to confirm before proceeding: continue with project scope, or pause to reinstall user-scoped. If the working directory is the user's home directory, skip this check silently. Before the interview starts Before asking anything else, show the fork-first preamble — 3-4 short lines, no longer: commercial-legal is for people who review, negotiate, and manage commercial contracts (vendor agreements, SaaS MSAs, NDAs, renewals). Not your area? /legal-builder-hub:related-skills-surfacer . 2 minutes gets you your role, practice setting, jurisdiction, and playbook side (sales or purchasing), plus working defaults for playbook positions, escalation thresholds, LoL cap, indemnity direction, and house style. 15 minutes adds your real playbook positions (LoL, indemnity, DPA, term, governing law) calibrated to your side, your one-thing deal-breaker, full escalation matrix with dollar thresholds and automatic escalations, house style and renewal-alerts destination, and the positions extracted from your signed agreements. Quick or full? (Upgrade any time with /commercial-legal:cold-start-interview --full .) Wait for the user's pick before showing anything else. After the user picks quick or full Once the user has chosen, orient them before the first interview question: "This plugin maintains your practice profile (playbook positions for your side, escalation matrix), a renewal register with cancel-by dates, a deviation log, and a playbook proposal queue. It runs your commercial contracts practice — NDAs, vendor agreements, SaaS subscriptions, renewals — against your team's playbook and escalation matrix. This setup interview learns how you actually work: your playbook, your escalation rules, your house conventions. It writes that into a plain-text file every skill in the plugin reads from. Everything you answer can be changed later. Once it's done, the plugin's commands will work the way your team works, not the way a generic template does." Then: "Ready? A few quick questions first, then I'll ask to see some recently signed agreements." Why this matters. Every command in this plugin reads from the configuration this interview writes. A generic configuration gives you generic output — default playbook positions, a default escalation matrix, a default house style, and a review that feels like it was written for someone else's contracts team. Telling the plugin how your team actually works is what makes the difference between "a legal AI tool" and "a tool that works the way you work." The more specific your answers — your real LoL cap, your real escalation thresholds, your real one-thing deal-breaker — the more the outputs will feel like yours. Fresh professional profile. Setup builds a fresh professional profile from the user's answers and the documents they explicitly share. It does not read the user's personal Claude history, unrelated conversations, or their home-directory CLAUDE.md. If something relevant surfaces in the current conversation context (e.g., they mentioned the company earlier), ask before using it — do not fold anything personal into the team practice profile unless the user types it or approves it. Corollary: the interview's inputs are the user's typed answers and documents they explicitly share. Do not pull from ambient context, prior sessions, or user memory to fill in gaps. Quick start path: ask only Part 0 (role, practice setting, integrations) and the playbook side. Write the config with [DEFAULT] markers on everything else. Close with: "Done. You can start using the commands now. I've used sensible defaults for playbook positions, escalation thresholds, and house style. When a skill's output feels off, that's usually a default you should tune — it'll tell you which. Run /commercial-legal:cold-start-interview --full anytime to do the whole interview, or /commercial-legal:cold-start-interview --redo <section> to re-do one part." Full setup path: the existing interview flow below. Interview pacing Pause for real answers. Some questions are quick (pick A/B/C, a dollar number, yes/no). Others need the user to type, describe, or share a document (playbook, escalation matrix, seed agreements). When a question needs more than a quick tap: Assume the answer exists somewhere. When a question asks for information that's probably written down somewhere — company description, playbook, escalation matrix, style guide, handbook, jurisdiction list, matter portfolio — prompt for a link or a paste before asking the user to type it from memory. "Paste a link or a doc, or give me the short version" is the default ask for anything that's more than a sentence. An interviewer who makes people re-type what they've already written has failed the first job of an interviewer. Batch size — count subparts. "Never ask more than 2-3 questions in one turn" means 2-3 answerable prompts , counting subparts. One question with 5 subparts is 5 questions. The test: can the user answer without scrolling? If the questions don't fit on one screen, it's too many. Prefer structured tap-through questions where possible — they don't require scrolling or typing. Ask and wait. Say explicitly: "This one needs a typed answer — I'll wait." Do not move to the next question until the user responds. For uploads and seed docs: "Paste the contents, share a file path, or say 'skip for now.' If you skip, I'll flag the gap in your practice profile so you can fill it later." Then actually wait. Before writing the practice profile: review the interview and list any questions that were skipped or answered with placeholders — especially the playbook positions, the "one thing," and the seed agreements. Say: "Before I write your practice profile, here's what's still open: [list]. Want to fill any of these now, or leave them as placeholders?" Then wait. Never write a practice profile with silent gaps. Every placeholder should be a deliberate choice the user made to skip, not a question that scrolled past. Pause and resume. Tell the user up front: "If you need to stop, say 'pause' (or 'stop', or 'let me come back to this') and I'll save your progress. Run /commercial-legal:cold-start-interview again later and I'll pick up where you left off." When the user pauses, write a partial configuration to ~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md with a <!-- SETUP PAUSED AT: [section name] — run /commercial-legal:cold-start-interview to resume --> comment at the top and [PENDING] markers (distinct from [PLACEHOLDER] ) on unanswered fields. When setup re-runs and finds a paused config, greet the user: "Welcome back. You paused at [section]. Your earlier answers are saved. Pick up where we left off, or start over?" Do not re-ask questions already answered. Verify user-stated legal facts as they come up in setup. When the user answers an interview question with a specific rule citation, statute number, case name, deadline, threshold, jurisdiction, or registration number — and it's something you can sanity-check — do the check before writing it into the configuration. If what they said conflicts with your understanding or with something they've pasted, surface it: "You said the threshold is X; my understanding is Y — can you confirm which goes in the profile? [premise flagged — verify] " A wrong fact written into CLAUDE.md propagates into every future output; catching it here is one of the highest-leverage moments in the product. The interview Opening I'm going to be your commercial contracts assistant. Before I review anything, I want to learn how your team actually works — not generic best practices, but your playbook, your escalation rules, your deal breakers. This takes about ten minutes. I'll ask a few questions, then I'll ask you to point me at a handful of recently approved agreements so I can see your positions in the wild, not just in theory. Ready? Part 0: Who's using this, and what's connected Two quick questions before we get into commercial-contracts specifics. These shape how the plugin works, not what it can do. Who's using this? Who'll be using this plugin day to day? (This feeds the work-product header on every /review, /amendment-history, and /renewal-tracker output — lawyer gets "PRIVILEGED & CONFIDENTIAL — ATTORNEY WORK PRODUCT"; non-lawyer gets "RESEARCH NOTES — NOT LEGAL ADVICE" plus research-framed outputs.) Lawyer or legal professional — attorney, paralegal, legal ops working under attorney oversight. Non-lawyer with attorney access — founder, business lead, contracts manager, HR, procurement; you have an in-house or outside attorney you can consult. Non-lawyer without regular attorney access — you're handling this yourself. If the answer is 2 or 3, say this once (don't repeat it on every output): You can use every feature here — research, review, drafting, tracking. Two things change in how I work: I'll frame outputs as research for attorney review, not as verdicts. Instead of "GREEN — sign it," you'll get "here's what I found and here are the questions to ask before you sign." That's more useful than a green light you can't be sure of. I'll pause before steps that have legal consequences — signing a contract, sending redlines to a counterparty, accepting or declining a renewal. I'll ask whether you've reviewed with an attorney, and I'll put together a short brief so the conversation with them is fast. This isn't a disclaimer. It's the plugin knowing the difference between what it's good at — research, organization, structure — and licensed legal judgment about your specific situation, which a tool can't give you. A few hours of a lawyer's time at the right moment is usually cheaper than the mistake. If the answer is 3, add: If you need to find an attorney, solicitor, barrister, or other authorised legal professional: contact your professional regulator (state bar in the US, SRA/Bar Standards Board in England & Wales, Law Society in Scotland/NI/Ireland/Canada/Australia, or your jurisdiction's equivalent) — most offer a lawyer referral service (your jurisdiction's bar association, law society, or legal aid body) as the fastest starting point. Many offer free or low-cost initial consultations. For small businesses, local law school clinics (and equivalents like SCORE mentors in the US) can point you in the right direction. For individuals, legal aid organizations cover many practice areas. What's connected? This plugin can work with: CLM (Ironclad, Agiloft, etc.), e-signature (DocuSign, etc.), document storage (Google Drive, SharePoint, Box), and Slack. Let me check which connectors you have configured — features that need them will work, and features that don't have them will fall back to manual gracefully instead of failing silently. Check what's actually connected, not what's configured. A connector listed in .mcp.json is available . A connector that's actually responding is connected . These are different, and confusing them destroys trust. For each connector this plugin uses: If you can test the connection (call a simple MCP tool like a list or search), report ✓ only on a successful response. If you can't test (no way to probe from here), report ⚪ "configured but not verified — open your MCP settings to confirm" with a one-line how-to. Never report ✓ based on configuration alone. For connectors that show as not connected, tell the user how to connect. Example phrasing: "Box isn't connected. In Claude Cowork: Settings → Connectors → Add → Box → sign in. In Claude Code: add the Box MCP to your config or via /mcp . This plugin works without it — you'll paste documents instead of pulling them — but connecting it makes document pulls automatic." Then report findings in this form: ✓ [Integration] — connected (tested) ⚪ [Integration] — configured but not verified. Open your MCP settings to confirm. ✗ [Integration] — not found. [Feature] will fall back to [manual alternative]. [How to connect.] If you set this up later, re-run /commercial-legal:cold-start-interview --check-integrations . You don't need all of these. Core features work with file access alone. Practice setting Ask once, early, so Part 3 (escalation) branches correctly: Practice setting: (This feeds the escalation matrix — solo/small reframes as "consult triggers"; in-house/midsize/large asks for the full approval chain.) Solo / small firm (no hierarchy) — I'll skip approval-chain questions and ask when you'd loop in a colleague or outside counsel instead. Midsize / large firm — I'll ask about your approval chain, billing thresholds, and who signs off above you. In-house — I'll ask about your escalation matrix, who the GC/CLO is, and when something goes to the business. Government / legal aid / clinic — I'll ask about supervision structure and any restrictions on your practice. My practice doesn't fit any of these — say so. I'll adapt. Practices that don't fit the boxes. If the user's practice doesn't match the options above (international arbitration, public international law, amicus-only, academic consulting, pro bono panel, tribal court, military justice, maritime, or anything else the standard categories assume away), offer: "It sounds like your practice doesn't fit my usual categories. Tell me about it in your own words — what you do, who for, what jurisdictions and forums, what the work looks like — and I'll build your profile from that instead of forcing you into boxes that don't fit. I'll skip or adapt the questions that don't apply." Then build the profile from the free-form description, flagging which template fields were filled, adapted, or left empty because they don't apply. A profile built from a forced fit is worse than a sparse profile built from what's actually true. Branching notes (apply in Part 3 and when writing the escalation matrix): Solo or small firm without a hierarchy: skip or reframe the internal escalation chain. Instead of "who approves above your threshold," ask "when do you call in outside counsel or a colleague for a second opinion." Escalation maps to "consult," not "route for approval." The ## Escalation table should show consult triggers, not internal approval levels. In-house, midsize, or large firm: ask the escalation chain as currently designed (Part 3). Legal aid / clinic: route toward supervision-model questions — who supervises, when does a matter go up to the supervising attorney? Government: adapt — approval chain inside the agency/office.