Sales Proposals skill for Codex
Win the deal without losing the margin. Codex writes proposals from discovery notes, restating the buyer's problem and success criteria and mapping each criterion to an approach and a deliverable. Pricing comes as clear options with an ROI section built on the buyer's own numbers. The SOW then protects delivery with testable acceptance criteria, customer responsibilities, a payment schedule and written change control, and nothing is sent without your confirmation.
curl -fsSL codexskills.dev/install.sh \ | sh -s -- sales-proposals Installs to ~/.agents/skills/sales-proposals/ for every project on this machine.
curl -fsSL codexskills.dev/cli.mjs -o codex-skills.mjsnode codex-skills.mjs install sales-proposals The CLI also searches, updates and removes skills. Run it with no arguments to pick from a list. Needs Node 20.12+.
curl -fsSL codexskills.dev/install.sh \ | sh -s -- --project sales-proposals Installs to .agents/skills/sales-proposals/ in the current repository. Commit it and everyone who runs Codex there gets it.
curl -fsSLO codexskills.dev/downloads/sales-proposals.zipunzip -o sales-proposals.zip -d ~/.agents/skills/ The archive holds the skill folder with its generated agents/openai.yaml. Unzip into .agents/skills/ to scope it to a repo.
& ([scriptblock]::Create((irm codexskills.dev/install.ps1))) sales-proposals PowerShell 5.1 or 7. Installs to %USERPROFILE%\.agents\skills\. Add -Project for the current repo.
104-line POSIX script. No sudo, no telemetry. Read install.sh
Use it in Codex
Start a prompt with $sales-proposals, or pick it from /skills.
You can also just describe the task: Codex reads the description below and loads the playbook when your request matches.
What Codex reads before loading it
Writes B2B sales proposals and statements of work that win and protect delivery - discovery-based executive summary in the buyer's terms, solution mapped to their success criteria, pricing options, ROI with stated assumptions, and SOWs with deliverables, acceptance criteria, customer responsibilities, milestones, payment schedule and change control. Use when asked to write a proposal, quote narrative, RFP response section, SOW or scope of work, or to review a draft proposal before it goes to a prospect.
Use it when
- A qualified prospect asked for a proposal or quote
- A deal is won and needs an SOW that delivery can live with
- Responding to sections of an RFP
- Projects keep overrunning because scope was never pinned down
Use something else when
- You are pitching an internal project or budget request: use proposal-writing
- You are a solo freelancer setting up quotes, contracts and invoicing: use freelance-operations
- You are evaluating vendors' proposals as the buyer: use vendor-evaluation
How to check the result
Every buyer success criterion maps to a deliverable with objective acceptance criteria, pricing arithmetic adds up, and all non-standard terms are flagged for internal approval.
The mistake it prevents
A fixed-fee SOW with vague deliverables and no change-control clause, which turns every request into free work.
A winning proposal reads like it was written for one buyer: it restates their problem and success criteria in their words, shows exactly how the offer meets each one, and makes the decision easy with clear pricing and next steps. A good SOW then makes the promise deliverable: every deliverable has acceptance criteria, every assumption is written down, and any change has a priced path.
Inputs (from discovery)
Do not draft without these; ask for call notes, emails, CRM fields or RFP text:
- Buyer’s problem and its cost (time, money, risk) in their words
- Success criteria: how they will judge the project in 6-12 months
- Decision process: who decides, who signs, who influences, procurement steps, date
- Budget range or pricing signals; competing options including “do nothing”
- Constraints: deadlines, security or compliance reviews, integration requirements
- Your delivery capacity and the standard terms (MSA, rate card, payment terms)
Missing items become explicit questions, not invented assumptions.
Proposal structure
| Section | Content | Length |
|---|---|---|
| Executive summary | Their situation, the outcome they want, your recommendation, price, timeline | Half a page; the only part some approvers read |
| Understanding | Problem, impact, success criteria (numbered, to reference later) | Half a page |
| Approach | How you will reach each success criterion; phases | 1-2 pages |
| Scope and deliverables | What they receive; what is out of scope | Table |
| Timeline | Milestones with dates or durations from start | Table or simple chart |
| Team | Named roles and relevant experience | Short |
| Investment | Pricing options, payment terms, what is included | Table |
| Value / ROI | Quantified benefit with assumptions they can check | Short |
| Why us | Proof tied to their criteria: case studies, references | Short |
| Assumptions and terms | Dependencies, validity period, reference to MSA | List |
| Next steps | Exactly what happens to start, with dates | 3-5 bullets |
Map every success criterion to an approach section and a deliverable; a criterion with no matching deliverable is a gap the buyer will notice.
Executive summary pattern
<Company> wants to <outcome> by <date> because <business driver>. Today <problem and
its cost, quantified>. We recommend <solution in one sentence>, delivered in <N> weeks
for <price or range>. Based on <assumption you can check>, this <benefit, quantified>.
To start on <date>, we need <decision/signature> by <date>.
Pricing options
Offer two or three options when the buyer’s priorities are uncertain; one when the scope is clear and agreed.
| Model | Use when | Protect yourself with |
|---|---|---|
| Fixed fee | Scope is well defined; buyer wants budget certainty | Tight scope, acceptance criteria, change control |
| Time and materials | Scope uncertain or discovery-heavy | Estimate range, not-to-exceed cap if requested, weekly burn reports |
| Phased (fixed discovery, then fixed or T&M build) | Large or ambiguous projects | Discovery deliverable that defines phase 2 |
| Retainer / subscription | Ongoing work | Defined hours or capacity, rollover rules, notice period |
| Value-based | Benefit is large and measurable | Clear baseline and measurement method |
Options table:
| | Essential | Recommended | Extended |
|---|---|---|---|
| Scope | Criteria 1-2 | Criteria 1-4 | 1-4 + training and 3 months support |
| Timeline | 6 weeks | 10 weeks | 10 weeks + support |
| Investment | $48,000 | $76,000 | $92,000 |
State currency, taxes (excluded or included), expenses policy and the proposal’s validity period (for example 30 days).
ROI section
Show the arithmetic with the buyer’s numbers and conservative assumptions:
Support agents spend ~6 hours/week on manual invoice lookups (your estimate, call on 10-14).
12 agents x 6 h x 46 weeks x $38/h loaded cost = $125,856/year.
If the integration removes 70% of that work: ~$88,000/year saved.
Payback on the Recommended option: ~10 months.
Label every input as theirs, yours, or an industry estimate. Do not inflate; the buyer’s finance team will check it.
Statement of work
# Statement of Work <number> under MSA dated <date>
1. Objectives - <linked to proposal success criteria>
2. Scope - in scope: <list>; out of scope: <list>
3. Deliverables and acceptance criteria
| # | Deliverable | Acceptance criteria | Due |
|---|---|---|---|
| D1 | Invoice lookup API integration | Returns invoice for 99% of test set of 200 order IDs in < 2 s; passes customer UAT script v1 | Week 6 |
4. Acceptance process - customer reviews within 5 business days; deemed accepted if no
written rejection listing non-conformities; one rework cycle included
5. Customer responsibilities - <access, data, SMEs, environments, decisions by dates>
6. Assumptions - <volumes, environments, third-party availability>
7. Timeline and milestones
8. Fees and payment schedule - <milestone or monthly; invoice terms>
9. Change control - changes requested in writing; provider returns impact on scope, cost
and schedule within 3 business days; no work on changes until signed change order
10. Project governance - contacts, status cadence, escalation path
11. Term and termination - per MSA unless stated
Acceptance criteria must be objective and testable. “Customer satisfied” is not acceptance; “passes UAT script v1 with no severity-1 defects” is.
Reviewing a draft
- Executive summary works alone and names price, timeline and next step
- Every success criterion maps to a deliverable
- Pricing arithmetic adds up; options differ in clearly described ways
- Out-of-scope list addresses the obvious “I assumed that was included” items
- Customer responsibilities and dependencies are explicit
- No promise the delivery team has not confirmed (dates, SLAs, integrations)
- Consistent with the MSA; any deviations flagged for legal or finance review
- Buyer’s name, logo and terminology correct everywhere (copy-paste from the last proposal is a classic error)
Gotchas
- Feature-dumping: pages about your company before the buyer’s problem appears.
- Unlimited revisions or “ongoing support” with no definition becomes free work.
- Fixed fee with vague scope transfers all risk to you; T&M with no estimate transfers it all to the buyer and stalls approval.
- Pricing in a proposal can form part of a binding offer depending on wording and law; include validity and “subject to MSA” language, and have legal approve new terms.
- Sending the proposal is outward-facing: prepare the document and email draft, but send only after the user confirms.
Output
Proposal (markdown or .docx via docx-documents), SOW draft, a criteria-to-deliverable traceability table, open questions for the buyer, and items requiring internal approval (non-standard terms, discounts, delivery commitments).
What the zip contains
SKILL.md plus the interface file below. Nothing runs on install.
agents/openai.yaml
Gives the Codex app a display name, a one-line description and a starter prompt for this skill.
interface:
display_name: "Sales Proposals"
short_description: "Win the deal without losing the margin"
default_prompt: "$sales-proposals Write a proposal for Acme's invoice integration project from these discovery call notes" About this skill
Written and maintained by the Codex Skills editors to the same editorial standard as every skill here: a selective description, a real workflow, known pitfalls and a way to verify the work. First published , last updated . MIT licensed. Something wrong or out of date? Tell us.