# Skill gallery · Sales as Code Skill gallery # Nine skills the AI roles run, and the work they produce Inside the app, each role runs skills like these on a schedule and stages the result for your approval. These nine need nothing connected: no CRM, no inbox, no calendar. Each card shows a sample written from Arcwright Labs, a made-up company in our demo workspace. ## Try one on your own site Messaging variants: we read up to four pages of your site and draft three sets of hook, value line and ask, with every claim tagged as on your site or an assumption to confirm. No account. We use your email only to limit free runs: we keep a one-way code made from it for two days, never the address, and we do not add it to a list. Free runs share a small daily budget that resets at 00:00 UTC. For a fuller read of your company, with buyers, competitors and a sample post, run the [free audit](https://salesascode.com/audit). ### Your messaging variants A draft from your public pages only. Confirm every line tagged as an assumption before you use it. ## The gallery GTM strategist ### Messaging variants Drafts three sets of hook, value line and ask from how you describe yourself. Every claim is tagged as backed by something you published or as an assumption for you to confirm. Reads in the app: Your business brief and the customer stories you approved. See a sample Sample. The Arcwright Labs brief and its approved proof. Pain in play: the Thursday forecast is built by hand from four spreadsheets. [assumption: confirm with two customers] Variant A Hook: Your forecast takes two days to build and is stale by Friday. Value: Cadence OS reads what reps do in the CRM, the calendar and on calls, and grades the forecast call before the meeting. [proof: Trailhead, a two-day roll-up to a same-morning number] Ask: Send last quarter's forecast and we will show which deals it got wrong. Variant B Hook: Your reps have nine deals and cannot say which one is moving. Value: A manager sees the deal that is quietly slipping, not the one that came up in the one-on-one. [from your site] Ask: Fifteen minutes on one slipping deal of yours. What is assumed: that buyers count the forecast in hours. Confirm before you use variant A. Try this one on your site GTM strategist ### Two-week content calendar Plans two weeks of posts: a theme per week, the format of each post and its hook, tied to what you sell. No invented statistics in the hooks. Reads in the app: Your business brief and the customer stories you approved. See a sample Sample. The Arcwright Labs brief. Week 1. Theme: the Thursday forecast Mon, question post: how long does your forecast take to build? No numbers of ours. Wed, carousel: the four spreadsheets a forecast is built from, and what each one misses. Fri, customer story: Trailhead's move to a same-morning number. [proof: Trailhead] Week 2. Theme: coaching the deal that is slipping Tue, post: why two call reviews a month cannot tell a manager which deal is in trouble. Thu, short video: a manager's Monday, before and after. GTM strategist ### Launch readiness check Checks a launch against its assets, messaging, targets and operations, and lists what blocks it, most serious first. Reads in the app: Your business brief, your approved proof and the accounts you add. See a sample Sample. The Arcwright Labs launch of its Cadence Coach add-on. Launch: Cadence Coach add-on Blocking: no page says where call data is stored, and "where does our call data go" is one of the four objections you hear most. Blocking: the price page shows 340 per seat per year but not that it is sold only with Cadence OS. Gap: no customer story for coaching yet. Launch on the Cadence OS proof and say so. Ready: positioning, the list of current customers to tell first, the announcement draft. Sales development ### Account research brief A one-page brief on one company: what they do, why now, the likely pain and the best way in. Reads in the app: The account you add and any signals saved against it. See a sample Sample. Northwind Analytics, an invented account in the Arcwright Labs book. Northwind Analytics. Data infrastructure, about 340 people. Why now: hired a CRO from a Cadence OS customer three weeks ago. A new CRO in the first 90 days is a trigger in your brief. Likely pain: two RevOps roles open in Denver and Toronto. The reporting may be getting rebuilt by hand. [assumption: confirm] Best way in: the new CRO already knows what a graded forecast looks like. Ask what they had before and what they miss. Who: VP Revenue Operations first. The CRO signs. Account executive ### Sixty-second pre-call brief The single goal of the call, the last touch, the objection most likely to come up and the ask. Reads in the app: The account and opportunity you saved, and your answer library. See a sample Sample. Quillon Insurance Tech, an invented account in the Arcwright Labs book. Quillon Insurance Tech. Proposal stage, 240k. Goal: agree who signs off the security questionnaire, and by when. Last touch: security sent 84 questions. Nineteen are about model providers and data retention. Likely objection: "Where does our call data go?" Answer from the answer library, not from memory. Ask: thirty minutes with their security lead this week. Account executive ### Value framing brief A one-page case built on the buyer's own numbers. Every figure is theirs or a cited customer result; it never invents one. Reads in the app: The account you saved, its discovery notes and your approved proof. See a sample Sample. Northwind Analytics, an invented account in the Arcwright Labs book. Northwind Analytics. Negotiation, 165k. Their goal: forecast within 5 percent for two quarters running. [their words, discovery] Their pain: the Thursday forecast is built by hand from four spreadsheets. [their words, discovery] Proof: Trailhead went from a two-day manual roll-up to a same-morning number. [proof: Trailhead] Missing: what an hour of RevOps time costs them. Ask before you put a figure on the page. Account executive ### Solution overview Two or three slides: the problem, how you solve it, and a test for when you are not the right fit. The objection table always names a real limitation. Reads in the app: Your business brief and the account you choose. See a sample Sample. The Arcwright Labs brief and its "when to walk away" notes. Slide 1. The problem: the forecast is assembled by hand, and coaching happens on the wrong deal. Slide 2. How Cadence OS works: reads CRM, calendar and call activity, proposes the next step, grades the forecast call. Slide 3. When we are not the right fit: under 15 reps, no CRM of record, or a platform team already building the reporting. Objection: "Can we self-host?" No. That is a real gap, and we say so. Sales engineer ### Technical competitive teardown Where you win and lose against a named alternative on architecture, integration and security. Each claim carries proof or is marked as an assumption. Reads in the app: Your business brief, its competitor notes and the account you choose. See a sample Sample. Arcwright Labs against Forecastly, an invented competitor, for Northwind Analytics. Against Forecastly, for Northwind Analytics. Win: coaching. Forecastly is strongest on roll-up and weak on rep behaviour. [win-loss notes] Lose: a buyer who only wants a forecast roll-up has the simpler buy there. Say so. Security: every write-back is proposed, never silent. [from your security page] Decision criteria to check: CRM write-back, SSO, no data leaving the US. Partnerships ### Partner recruiting brief A profile of the partners worth recruiting, drawn from where your customers cluster, with the angle for each. It never invents company details. Reads in the app: Your business brief and the accounts you add. See a sample Sample. The Arcwright Labs brief and its buying triggers. Profile 1: RevOps consultancies that run CRM migrations for software companies of 80 to 600 people. Why: a CRM migration is one of your four buying triggers. Angle: forecast reporting gets rebuilt during a migration. Arrive before the platform team starts building it. Do not recruit: services-led firms that resell their own dashboards. They compete with you in deals under 50k. ## Not sure where to start? If pipeline is thin, answer six questions and find the layer that is broken: foundation, targeting, data, copy, sending or learning. [Find the broken layer](https://salesascode.com/diagnose) [Run the free audit](https://salesascode.com/audit) [See pricing](https://salesascode.com/pricing) --- More for agents: [llms.txt](https://salesascode.com/llms.txt), [full reference](https://salesascode.com/llms-full.txt), [OpenAPI](https://salesascode.com/openapi.json), [sitemap](https://salesascode.com/sitemap.xml).