Build a proposal that survives procurement
AI proposal generator is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
AI proposal generator is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
A proposal is read by people who were on none of the calls, so it has to open with what the buyer told you, in their words — get that visibly right and it buys the read for everything after it.
| name | proposal-structure |
|---|---|
| description | Our proposal skeleton and what each section must contain. Use when drafting a proposal, statement of work, or any document that asks a buyer to commit. |
| allowed-tools | none — this skill adds judgement, not reach |
| compatibility | Zealos, Claude Code, Cursor, Codex and any client that reads SKILL.md |
| license | Free to use and edit |
--- name: proposal-structure description: Our proposal skeleton and what each section must contain. Use when drafting a proposal, statement of work, or any document that asks a buyer to commit. --- # Proposal structure A proposal is read by people who were not on any of the calls. Write for the skeptic in the room, not for your champion. ## Sections, in order 1. **What you told us** — their situation and pain, in their words, with evidence from the calls. If this section is wrong, nothing after it matters. Getting it visibly right is what buys the rest of the read. 2. **What success looks like** — their metric, their number, their timeframe. Not ours. 3. **What we will do** — scope, in plain language. Phases with dates. 4. **What we need from you** — the part most proposals omit and every failed implementation traces back to. Access, people, decisions, and by when. 5. **What it costs** — plainly. One table. No "investment" euphemism. 6. **Why us** — last, and short. Two proof points relevant to *their* situation, not a logo wall. 7. **Next step** — one action with a date. ## Rules - Quote them. A line in their own words, early, does more than a page of ours. - Every claim gets a number or gets cut. - No section that exists to look thorough. - If a section could appear unchanged in a proposal to a different company, it is not doing any work. ## Length Shorter than they expect. Detail belongs in an appendix that the champion can forward to procurement.
The first draft arrives assembled from the actual calls rather than from last quarter's proposal with the company name swapped.
Scope stops being negotiated after signature, because the what-we-need-from-you section made the dependencies explicit while there was still leverage.
You stop spending a Sunday on every proposal, and the ones you send are shorter than the ones you used to agonise over.
What they told you, what success looks like, what you will do, what you need from them, what it costs, why you, and one next step. In that order.
Shorter than they expect. Detail belongs in an appendix the champion can forward to procurement without forwarding the argument.
Yes — that is what makes the first section credible. Pair it with a recap skill so the quotes are already extracted.
Post-call automation turns a recorded conversation into the three things that outlive it: the commitments each side made, the fields that changed, and the follow-up that has to go out today.
/call-recap-formatA pre-call brief is worth writing only if it changes what you ask: who is in the room, what they are measured on, what happened last time, and the one question you cannot afford to leave without answering.
/discovery-questionsAI email triage is worth having only when it sorts by consequence rather than by recency: what blocks a deal, what someone is waiting on you for, and what can be read at the end of the week.
/email-rulesCRM data hygiene survives AI only if one rule is written down first: a field left blank is a known unknown, while a field filled in by inference is an unknown unknown that will be reported on as fact.
/crm-field-conventionsQualification is a claim about evidence, not a score: a deal is qualified when every criterion has a source you can quote, and unevidenced criteria are named rather than filled in optimistically.
/meddiccA writing-voice skill is the highest-leverage thing you can give an AI, because it changes every sentence the product produces and it needs no tools, no integrations and no data.
/our-writing-voice