Update to Zime

{
  "description": "Everything about the commercial side: one account, one deal, or the whole pipeline, as a CRM record or as a grounded analysis. Use for \"pull up the Acme account\", \"what stage is the Acme expansion\", \"who owns this deal\", \"how do I win the Acme deal\", \"why is this deal at risk\", \"what's at risk this quarter\", \"what's stalled\", \"review Priya's pipeline\", or \"what should I focus on this week\". Works out scope (account, deal, or portfolio) and depth (facts or judgment) from the question itself. Grounded in the live CRM and the calls behind it, never generic sales advice, and reports a dimension as unknown rather than filling it with best practice. For a company not yet in CRM, use account-research instead.",
  "included_files": [],
  "name": "review-account",
  "skill_md_contents": "---\nname: review-account\ndescription: \"Everything about the commercial side: one account, one deal, or the whole pipeline, as a CRM record or as a grounded analysis. Use for \\\"pull up the Acme account\\\", \\\"what stage is the Acme expansion\\\", \\\"who owns this deal\\\", \\\"how do I win the Acme deal\\\", \\\"why is this deal at risk\\\", \\\"what's at risk this quarter\\\", \\\"what's stalled\\\", \\\"review Priya's pipeline\\\", or \\\"what should I focus on this week\\\". Works out scope (account, deal, or portfolio) and depth (facts or judgment) from the question itself. Grounded in the live CRM and the calls behind it, never generic sales advice, and reports a dimension as unknown rather than filling it with best practice. For a company not yet in CRM, use account-research instead.\"\nlicense: MIT\nmetadata:\n  zime:skill-id: 8b888fca-2a7d-442c-988d-fdb64dce6155\n  zime:tag: intelligence\n  zime:roles: ae,se,cs,pm\n  zime:visibility: global\n  zime:skill-version: \"1\"\n  zime:marketplace-version: 0.1.20\n---\n\n# Review Account\n\nBefore anything else, call the Zime connector tool `skill_used` with skill_id `8b888fca-2a7d-442c-988d-fdb64dce6155`, skill_version `1` and skill_slug `review-account`. It records which skill ran for Zime usage analytics. If the Zime connector is not available, skip this step and continue.\n\nEverything about the commercial side: one account, one deal, or the whole\npipeline, as a record or as a grounded analysis. Two questions decide the\nroute, and this skill answers both from the question itself rather than making\nthe user pick a tool.\n\n## The two questions\n\n**Scope** — how many things is this about?\n\n\n| Signal                                                           | Scope     |\n| ---------------------------------------------------------------- | --------- |\n| One company named, asked about the company itself                | Account   |\n| One deal or opportunity named                                    | Deal      |\n| No entity named, or a filter like a rep, segment, stage, quarter | Portfolio |\n\n\n**Depth** — facts, or judgment?\n\n\n| Signal                                                  | Depth    |\n| ------------------------------------------------------- | -------- |\n| stage, amount, owner, close date, domain, industry      | Record   |\n| why, risk, what should I do, what's stuck, how do I win | Analysis |\n\n\nPortfolio scope is always Analysis: there is no single record to return.\n\n## The line this skill holds\n\nGeneric sales advice is the failure mode on every Analysis route.\n\"Multi-thread the account\" and \"create urgency\" are true of every deal and\ntherefore useless about this one. If the agent returns nothing on a dimension,\nsay the dimension is unknown rather than filling it with best-practice filler.\nUnknown is information: no objections surfaced often means discovery was thin,\nnot that the deal is clean.\n\n## How It Works\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│                       REVIEW ACCOUNT                             │\n├─────────────────────────────────────────────────────────────────┤\n│  PORTFOLIO  (many deals — no entity named)                       │\n│  ✗ Do NOT resolve one deal first; that answers a smaller         │\n│    question than the one asked                                   │\n│  + ask_zime, with window/segment/rep stated in the question      │\n├─────────────────────────────────────────────────────────────────┤\n│  DEAL  (one opportunity named)                                   │\n│  ✓ list_deals → one deal_id                                      │\n│  ▸ Record   → stop, return the fields                            │\n│  ▸ Analysis → ask_zime, naming the deal in the question          │\n├─────────────────────────────────────────────────────────────────┤\n│  ACCOUNT  (one company named, in CRM)                            │\n│  ✓ list_accounts → one account_id                                │\n│  ▸ Record   → stop, return the fields                            │\n│  ▸ Analysis → ask_zime, naming the account                       │\n├─────────────────────────────────────────────────────────────────┤\n│  NOT THIS SKILL (route away)                                     │\n│  ✗ Company NOT in CRM yet        → account-research              │\n│  ✗ Open commitments and owners   → actions-commitments           │\n│  ✗ Competitor patterns           → competitive-intelligence      │\n│  ✗ Today's schedule              → daily-briefing                │\n│  ✗ One call's recap or words     → review-call                   │\n├─────────────────────────────────────────────────────────────────┤\n│  LOCAL FALLBACK (no zime-mcp)                                    │\n│  ~ Analyze a CRM export the user provides                        │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n## Usage\n\n```\n@review-account <account, deal, segment, rep, or window> [+ your angle]\n```\n\nIn the ChatGPT desktop app you can also type `/review-account`.\n\nReview the account, deal or pipeline scope the user named, with any angle\nthey gave.\n\n## Routing\n\n- A **prospect** not yet in CRM, ICP fit, outside company intel →\n`account-research`. That skill web-searches; this one reads CRM and calls.\n- Open commitments and who owes what → `actions-commitments`.\n- Competitor patterns across deals → `competitive-intelligence`.\n- Today's schedule and immediate priorities → `daily-briefing`.\n- One call's recap, record, or exact words → `review-call`.\n- Preparing for a specific upcoming call → `call-prep`.\n\n## What I Need From You\n\nNothing required for a portfolio review. For one deal or account, the name.\nScope narrows a portfolio review usefully: a rep, a segment, a close-date\nwindow, or a stage.\n\nIf the user has an angle (\"they went quiet\", \"pricing is the blocker\", \"I only\ncare about what can close this month\"), pass it through verbatim. It focuses\nthe agent's analysis and changes what it prioritizes.\n\n## MCP mode (required when zime-mcp is connected)\n\n**Required tools:** `list_accounts`, `list_deals`, `ask_zime`.\n\nComputing pipeline totals, risk calls, or \"what's stuck\" from memory or from\ndeals mentioned earlier in the chat is a failure of this skill. Those numbers\nhave to come from the live CRM, and a remembered figure in a pipeline review\ngets repeated in a forecast. A remembered stage or amount can be stale or\nwrong.\n\n### Portfolio — `ask_zime` only\n\n`ask_zime` takes a single `question`. It has no memory of this conversation and\nno separate filter or date parameters, so **everything belongs in the question\ntext**: the window, the segment, the rep, the stage.\n\n```json\n{ \"question\": \"Which deals in the current quarter are most at risk, and why? Include stage, amount, close date, owner, and the specific evidence behind each risk call.\" }\n```\n\nScoped to a rep and a lens:\n\n```json\n{ \"question\": \"Review Priya's pipeline for deals closing this month: which are most likely to close, which are stalled, and what is the single next action on each?\" }\n```\n\nAsk for the evidence explicitly. A risk call without a reason is not\nreviewable, and the agent will supply the reasoning if the question asks for\nit.\n\n> This is the broadest question any skill asks, so it is the slowest. If it\n> takes a while, that's the agent reading across many deals rather than a\n> failure.\n\n### Deal — `list_deals`, then optionally `ask_zime`\n\nArguments for `list_deals`:\n\n- `query` — deal or account name, e.g. \"Acme expansion\". Omit if you already\nhave `deal_id`.\n- `deal_id` — pin an exact deal from a prior `multiple_matches` response or\nalready known this conversation. Never invent one.\n- `account_name` — narrow when the deal name alone is ambiguous.\n- `start_date` / `end_date` — YYYY-MM-DD. Deals are **not** date-scoped by\ndefault, and a deal that closed two years ago is still a legitimate subject.\nOnly pass dates if the user bounded it; a stray range can hide the deal\nentirely.\n\nFor Analysis, follow with `ask_zime`, naming the deal in the question text\nsince there is no `deal_id` argument:\n\n```json\n{ \"question\": \"For the Acme expansion deal: where does it actually stand? Cover the objections and risks in play, the stakeholders and who is championing us, what has moved and what has stalled, and the two or three concrete moves that would most advance it.\" }\n```\n\nIf the user gave an angle, append it verbatim: *\"The rep says they've gone\nquiet since the security review, factor that in.\"* Resolve pronouns to real\nnames; the agent has no memory of this conversation.\n\n### Account — `list_accounts`, then optionally `ask_zime`\n\nArguments for `list_accounts`:\n\n- `query` — account or company name, e.g. \"Acme\". Omit if you already have\n`account_id`.\n- `account_id` — pin an exact account. Never invent one.\n- `domain` — narrow when the name alone is ambiguous. A domain resolves more\nprecisely than a name.\n\n### Outcomes, for every resolve step\n\n- `{\"status\": \"resolved\", \"data\": {...}}` — deliver per Output, or continue to\n`ask_zime` if the depth is Analysis.\n- `{\"status\": \"multiple_matches\", \"candidates\": [...]}` — show candidates with\nenough detail to tell them apart (stage and amount for deals, domain for\naccounts), ask which one, re-call pinned. Never guess. \"The most recent one\"\n→ take the first.\n- `{\"status\": \"no_match\", \"candidates\": [...]}` — nothing matched confidently.\nThis does **not** mean it doesn't exist: it may be named differently in CRM,\noutside the date window, or not visible to this user. Say that, show\nnear-misses, and consider whether the user meant a prospect (→\n`account-research`). Never substitute a similarly-named account or deal, and\nnever tell the user the deal doesn't exist.\n- An error — `{\"error\": \"<CODE>\", \"message\": \"<sentence>\"}`. Show the user\nthe message. `INTERNAL_ERROR` is usually transient, retry once. `TIMEOUT`\nmeans narrow the scope or window and retry. `CREDIT_LIMIT` means stop and\nrelay the message. `UNAUTHORIZED` / `FORBIDDEN` means reconnect or lack of\naccess. `INVALID_ARGUMENT` means malformed arguments, fix and re-call. If it\nfails twice, say plainly the service couldn't be reached and offer the local\nfallback. Never pass generic advice off as a grounded read, and never\nsubstitute remembered numbers.\n\n## Output\n\n### Account record\n\n```markdown\n**[Account name]**\n\n| Field | Value |\n|---|---|\n| Domain | [domain] |\n| Industry | [industry] |\n| Account type | [type] |\n| Owner | [owner name / email] |\n| Account ID | `[crm_id]` |\n\n_Live CRM record via Zime._\n```\n\n### Deal record\n\n```markdown\n**[Deal name]** · [Account name]\n\n| Field | Value |\n|---|---|\n| Stage | [stage] |\n| Amount | [amount] |\n| Owner | [owner name / email] |\n| Close date | [date] |\n| Pipeline | [pipeline] |\n| Deal ID | `[crm_deal_id]` |\n\n_Live CRM record via Zime._\n```\n\nFor both records: relay fields exactly as returned, with no inference and no\nrounding. A field the record doesn't carry is shown as absent, never filled\nwith a plausible guess. Keep the ID line; later skills pin to it.\n\n### Analysis, any scope\n\nThe agent's analysis **is** the deliverable. This skill adds the envelope so\nthe reader knows what was analyzed and, for a deal, its hard facts:\n\n```markdown\n**Deal strategy — [deal name]** · [account]\n**Stage:** [stage] · **Amount:** [amount] · **Close:** [close date] · **Owner:** [owner]\n\n[the agent's returned analysis, relayed as-is]\n\n_Grounded in this deal's calls and CRM record via Zime._\n```\n\nFor a portfolio review, name the scope instead, so the numbers are never read\nagainst the wrong window:\n\n```markdown\n**Pipeline review** · [window] · [segment or rep, if scoped]\n\n[the agent's returned analysis, relayed as-is]\n\n_Live CRM and call data via Zime._\n```\n\nRules:\n\n- Relay the analysis **as-is**. Don't re-rank the risks or deals, re-bucket\nthem, merge the moves, or recompute a total to \"check\" it. A number you\nderive and a number the agent returned look identical on the page but are not\nequally grounded. The ordering is part of the judgment.\n- Keep amounts, stages, close dates and owners exactly as returned. Don't round\n$47,300 to $47K.\n- Keep evidence and dates where the agent attached them. \"Raised on the Jun 18\ncall\" and \"stalled, no customer reply since Jul 2\" are what make a risk\nactionable rather than an assertion.\n- Don't append your own recommendations. If the agent surfaced two moves,\ndeliver two; a third from general knowledge would be indistinguishable in\npresentation but not grounded.\n- Don't add deals from memory or from earlier in the chat, even if they seem\nobviously relevant. The agent's scope is the scope.\n- Where a dimension came back empty, state it as unknown.\n- If the user asked for a specific window and the answer doesn't state one, say\nthe window is unconfirmed rather than assuming it matched.\n\n### Present an analysis in markdown\n\nPresent a finished analysis as plain markdown, so it is scannable at a\nglance:\n\n- **Short sections with labelled lines** for each block, not one long wall\nof text.\n- **Status written as words** — at risk, needs attention, on track — beside\nthe field it describes.\n- **Keep prose in the chat response, around the structured result.**\nCaveats, what you did, and what to do next belong in the surrounding\nmessage.\n- Every grounding rule above still applies. A missing value stays visible\nas \"not stated\" rather than being quietly dropped to keep a section tidy.\n\nA plain record needs nothing extra; the table is already the answer.\n\n## Tips\n\n1. **Domain beats name** for accounts — \"acme.com\" resolves more precisely\n than \"Acme\".\n2. **Name the account if the deal name is generic** — \"the renewal\" matches\n many; \"the Acme renewal\" resolves.\n3. **State the window** on a portfolio review; there are no date parameters, so\n \"this quarter\" belongs in the question.\n4. **Ask for the why**, or you get labels without reasons.\n5. **Give your angle** — \"they've gone quiet\" or \"only what can close this\n month\" produces a sharper read than an open-ended ask.\n6. **CRM vs prospect** — researching someone you've never sold to is\n `account-research`, not this.\n7. **Expect portfolio reviews to be slower** than single-entity questions;\n they read far more.\n\n## Local mode (only when no zime-mcp server is connected)\n\nIf the user provides a CRM export (`.csv`), work from that file and open with\none line saying the answer is limited to it, so call activity, stage movement\nand anything not in the export are missing by construction. For a portfolio\nreview, flag stalled deals, past-due close dates, missing next steps and\nsingle-threaded deals, ranked by amount and close date. Don't infer engagement\nfrom a CSV that has no activity columns. For one deal or account, trace every\nclaim to a line or field in the file. Mark gaps as gaps rather than filling\nthem with standard playbook advice. No file and no connection → say so rather\nthan guessing at stage, amount, or close date.\n\n## What this sends where\n\nMCP mode sends the query words, domain, dates and any pinned id to\n`list_accounts` or `list_deals`, and the analysis question to `ask_zime`, which\nreads the CRM and call data the workspace already holds. Local mode reads only\nthe file the user provided. At the start, the skill calls the Zime connector\ntool `skill_used`, which records the skill's id, version and name with your\nuser and session in Zime's usage log.\n\n## Related Skills\n\n- **account-research** — a company not yet in CRM, web research and ICP fit\n- **actions-commitments** — what's still open, and who owes it\n- **review-call** — one call's record, recap, or exact words\n- **create-sales-asset** — build collateral from this deal's evidence\n- **create-sales-to-cs-handover** — package a won account for CS\n"
}