Update to Zime

{
  "description": "Builds a sales asset from real customer evidence only, never made up facts. Covers account review decks (as a slide-by-slide outline), one-pagers, battlecards, case studies, and objection talking points. Always an internal draft, not customer-ready.",
  "included_files": [
    {
      "relative_path": "references/deck-grammar.md",
      "size_in_bytes": 5198
    }
  ],
  "name": "create-sales-asset",
  "skill_md_contents": "---\nname: create-sales-asset\ndescription: \"Builds a sales asset from real customer evidence only, never made up facts. Covers account review decks (as a slide-by-slide outline), one-pagers, battlecards, case studies, and objection talking points. Always an internal draft, not customer-ready.\"\nlicense: MIT\nmetadata:\n  zime:skill-id: 33b3e084-8df2-40bd-b91f-1d5199dd9a14\n  zime:tag: communication\n  zime:roles: ae,marketing\n  zime:visibility: global\n  zime:skill-version: \"1\"\n  zime:marketplace-version: 0.1.20\n---\n\n# Sales Asset Builder\n\nBefore anything else, call the Zime connector tool `skill_used` with skill_id `33b3e084-8df2-40bd-b91f-1d5199dd9a14`, skill_version `1` and skill_slug `create-sales-asset`. It records which skill ran for Zime usage analytics. If the Zime connector is not available, skip this step and continue.\n\nTurns \"build me an account review for Acme\" into a draft in our house format.\nThis skill owns the **structure and framing**; the agents own the **facts**. It\nnever originates a metric, a quote, or a \"customer said\" line.\n\n## The hard rule\n\nEvery factual claim in the asset — a metric, a quoted line, a stated customer\noutcome, a competitive win — must trace to something a tool returned in **this\nconversation**, with the account and date attached. If a needed fact wasn't\nreturned, the asset marks the gap open rather than filling it with a plausible\nnumber or a paraphrased \"customers typically say\" line.\n\nThis is the one rule the skill cannot bend. Arranging real tool output is the\njob; inventing supporting evidence is a failure of it — even when the invented\nline looks exactly like something a customer would say. An asset is the\ndeliverable most likely to be pasted into a customer-facing deck, which is\nprecisely why fabrication here is the most expensive.\n\n## How It Works\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│                   SALES ASSET BUILDER                            │\n├─────────────────────────────────────────────────────────────────┤\n│  STEP 1 — RESOLVE THE SUBJECT                                    │\n│  ✓ list_accounts (account asset) or list_deals (deal asset)      │\n├─────────────────────────────────────────────────────────────────┤\n│  STEP 2 — GATHER DATED EVIDENCE                                  │\n│  + ask_zime — signals, risks, expansion, outcomes, patterns │\n│  + get_transcript — the exact line, when a real quote is needed  │\n├─────────────────────────────────────────────────────────────────┤\n│  STEP 3 — ASSEMBLE (house format)                                │\n│  ✓ references/deck-grammar.md — the section order we use         │\n│  ✓ Every row carries evidence + date + account                   │\n├─────────────────────────────────────────────────────────────────┤\n│  STEP 4 — DELIVER                                                │\n│  ✓ Account review deck → slide-by-slide outline in markdown      │\n│  ✓ Every other asset type → markdown, ready to paste             │\n├─────────────────────────────────────────────────────────────────┤\n│  ALWAYS                                                          │\n│  ! Internal draft. Needs review before it leaves the building.   │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n## Usage\n\n```\n@create-sales-asset <asset type> for <account, deal, or competitor>\n```\n\nIn the ChatGPT desktop app you can also type `/create-sales-asset`.\n\nBuild the asset the user asked for.\n\n## Routing\n\n- A pipeline stat, risk or strategy on one deal, or one CRM fact, with no\n  asset to build → `review-account`.\n- A recap of one call, or the email after it → `review-call` / `follow-up`.\n- Competitive evidence gathering with no asset yet →\n  `competitive-intelligence` (then come back here to shape it).\n- Researching a **prospect** we haven't sold to → `account-research`.\n\n## What I Need From You\n\n**Which asset**, and **what it's about**:\n\n| Asset | Subject | Typically needs |\n|---|---|---|\n| Account review deck | one account | signals, risks, expansion, open items |\n| Proof-point one-pager | one account/deal | outcomes with numbers, a real quote |\n| Battlecard section | a competitor | how we position, objections, wins |\n| Case-study writeup | one customer | before/after, outcomes, a quote |\n| Objection talking points | an objection theme | how it was handled successfully |\n\nIf the user names an audience (\"for the CS team\", \"for the QBR\"), keep it —\nit changes emphasis, not evidence.\n\n## MCP mode (required when zime-mcp is connected)\n\n**Required tools:** `list_accounts` / `list_deals` (resolve), `ask_zime`\n(evidence), `get_transcript` (exact quotes).\n\n> Hand-building a proof point from general product knowledge while\n> `ask_zime` is available is a failure of this skill: only it sees the\n> workspace's real calls, CRM records, and extracted signals.\n\n### Step 1 — resolve the subject\n\n```json\n{ \"query\": \"Acme\" }\n```\n\n`multiple_matches` → show candidates, ask, pin. Never guess: building an asset\nabout the wrong account wastes the whole asset.\n\nSkip this step for a competitor-only battlecard — go straight to `ask_zime`.\n\n### Step 2 — gather evidence, asking for dates explicitly\n\nName the account or deal in the question itself, since `ask_zime` has no\nseparate id argument:\n\n```json\n{ \"question\": \"For the Acme account: what expansion signals, churn risks, and open commitments appear in our calls? For each one give the evidence, the call date, and who said it.\" }\n```\n\nCross-account or competitive:\n\n```json\n{ \"question\": \"How do we position against Initech in deals we have won, and what objections about them recur? Include the account and date for each example.\" }\n```\n\n**Always ask for the date and the speaker.** An undated claim can't go in an\nasset — see the deck grammar. If the agent returns claims without dates, ask\nonce more for them rather than shipping undated rows.\n\n### Step 3 — exact quotes only via `get_transcript`\n\nIf the asset needs something in quotation marks:\n\n```json\n{ \"query\": \"Acme renewal\", \"start_date\": \"2026-06-01\", \"end_date\": \"2026-06-30\" }\n```\n\nA quote in the asset must come from here. Putting an agent's paraphrase inside\nquotation marks is fabrication with extra steps.\n\n### Outcomes\n\nEach tool returns the usual shapes — `resolved` / `multiple_matches` /\n`no_match`, or `{\"error\": \"<CODE>\", \"message\": \"<sentence>\"}` (show the user\nthe message; `INTERNAL_ERROR` retry once; `TIMEOUT` narrow and retry;\n`CREDIT_LIMIT` stop and relay the message; `UNAUTHORIZED` / `FORBIDDEN` means\nreconnect or lack of access). If a needed evidence call fails or returns\nnothing, the corresponding asset section says the gap is open. Never\nsubstitute a remembered figure.\n\n## Output\n\nFollow [references/deck-grammar.md](references/deck-grammar.md) for section\norder and the evidence-table shape. The short version:\n\n```markdown\n# [Asset title] — [subject]\n_Internal draft. Assembled from Zime call and CRM data on [date]. Not reviewed\nby marketing or legal._\n\n## The problem (in their words)\n> \"[verbatim quote]\" — [name, title], [date]\n\n## [Evidence section — signals / risks / expansion]\n| [Account] | [Signal] | Evidence | Date | Next action |\n|---|---|---|---|---|\n| [account] | [signal] | [what was said] | [date] | [action] |\n\n## What this means\n[synthesis — clearly framing, not new facts]\n\n## Open gaps\n- [section with no supporting evidence returned]\n\n---\n_Sources: Zime call recordings — [account], [dates]._\n```\n\nPresent the finished asset as plain markdown: short sections with labelled\nlines, status written as words (at risk, needs attention, on track), any\ncommentary in the chat around it, and gaps kept visible as \"not stated\".\n\n## Deliver in the right format\n\n- **Account review deck:** this is a presentation, not a document. Break the\n  content above into one slide per section (title, problem, evidence, what\n  it means, gaps), as a markdown outline with one heading per slide, so the\n  user can paste it into Slides or another deck tool. Tell the user the\n  layout will need a quick pass in their deck tool to clean up spacing and\n  bullets.\n- **Every other asset type** (one-pager, battlecard section, case study,\n  objection talking points): these are documents, not slides. Hand back the\n  markdown, ready to paste into Docs or another editor.\n\n### Rules\n\n- **Every row: account, evidence, date.** A row missing any of the three\n  doesn't ship — that's the format's whole discipline.\n- **Quotes come from `get_transcript`**, attributed to a named speaker with a\n  date. No exceptions.\n- **Relay claims close to the agent's phrasing.** Don't sharpen \"the\n  integration worried them\" into \"integration is why we're losing\".\n- **Never invent a number.** No ROI figure, headcount, or percentage that a\n  tool didn't return. If the asset format wants an ROI section and no numbers\n  came back, the section says so.\n- **Keep Open gaps.** Removing it makes a partial asset look complete.\n- **Always caption it as an internal draft**, every time. Calls contain\n  confidential customer language; anything built here needs review before it\n  goes to a prospect, into an external deck, or anywhere published.\n\n## Tips\n\n1. **Name the asset type** — \"account review deck\" and \"proof point\" have\n   different section orders.\n2. **Ask for dates in the evidence question** — retrofitting them later means\n   re-calling.\n3. **Fetch quotes separately** — one `get_transcript` call is what makes a\n   quote quotable.\n4. **Gaps are useful** — they tell the rep what to go ask on the next call.\n\n## Local mode (only when no zime-mcp server is connected)\n\nIf the user provides transcripts or a CRM export, build a narrower asset from\nthose files only — same rule: every claim traces to a line or field in the\nprovided files. Open with one line saying the asset covers only those files,\nnot the full account history, and repeat the internal-draft caveat.\n\n## What this sends where\n\nMCP mode sends query words (to `list_accounts`/`list_deals`), question text\nnaming the subject (to `ask_zime`), and query/dates plus a `call_id`\n(to `get_transcript`). Local mode reads only the provided files. At the\nstart, the skill calls the Zime connector tool `skill_used`, which records the\nskill's id, version and name with your user and session in Zime's usage log.\n\n## Related Skills\n\n- **competitive-intelligence** — gather competitor evidence first\n- **review-account** — the analysis behind a deal-focused asset\n- **create-sales-to-cs-handover** — the handover doc, a different house format\n- **review-call** — the exact quotable line from one call\n"
}