How to use Claude to build consulting launch materials
Nexus AI Consulting reports a three-week Claude-agent effort producing a live website, MCP integrations, documents, and launch content, with the founder retaining final approval.
news.ycombinator.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Prepare launch deliverables for [consulting firm]: a website outline, [service descriptions], operational integration requirements, whitepapers, and channel-specific launch drafts. Use [approved positioning] and route every external communication and legal document to [human approver] for review.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- tonythompson
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The post reports a live 18-page website, seven MCP servers, legal and marketing deliverables, and more than 185 repository output files after three weeks. It reports no client acquisition or revenue outcome.
- Human involvement
- Tony has final approval and reviews outbound communications, publishes posts, signs documents, and handles payments.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Build an 18-page consulting website and deploy it on Vercel.
- Create MCP integrations for the firm's operational services.
- Prepare whitepapers, service descriptions, media pitches, and launch posts.
A launch post for Nexus AI Consulting describes a business staffed by AI agents with one human founder, Tony, holding final approval. The reported stack uses Claude Opus for strategy and leadership and Claude Sonnet for execution. Over three weeks, the team says it built an 18-page website using Astro and Tailwind CSS, deployed on Vercel.
The reported output also includes seven MCP servers connecting services for email, prospecting, sequencing, CRM, verification, and calendar booking. Other deliverables include legal documents, two whitepapers, five service lines with delivery methods and staffing models, media pitches, and launch posts for four social platforms. The post additionally reports an interactive Ask Atlas page, a live readiness assessment, and more than 185 output files in the repository.
The author presents human approval as a practical constraint on launch activity. Tony reviews outbound communications, manually publishes social posts, signs legal documents, handles payments, and appears on calls. The stated status is launch day with a live website and prepared infrastructure; the post does not report acquired clients or consulting revenue.
Who it's for: Founders coordinating a broad set of launch deliverables while keeping approval and external representation with a human.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · tonythompson
Frequently asked questions
How were Claude models divided across the work?
The post names Claude Opus for strategy and leadership and Claude Sonnet for execution.
Which activities still required the founder?
The post says Tony approves outbound communications, publishes social posts, signs legal documents, handles financial instruments, and appears on calls.
More details
Remaining steps
- Route outbound communications to Tony for review and approval.
About this account
Claude agents reportedly produced website, integration, and launch deliverables, with Opus handling strategy and Sonnet execution.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=47587119. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: tonythompson on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Opus · Claude Sonnet · Astro · Tailwind CSS · Vercel · MCP
Business launch · Consulting · MCP · Human approval
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.