by Audra Carpenter

Screen potential clients against your ideal-client criteria and log the scores in a spreadsheet

Have Claude Cowork screen potential clients against your ideal-client criteria and log each score with a link to its proof in a spreadsheet, the agent the author describes.

What the agent did

Source-reported example · as described by Soon Lai (@soonlaii), fractional CTO and founder of Troolla Tech, which sells AI and fractional-CTO services to founders (per his X bio and troolla.com)

The author, Soon Lai, a fractional CTO who runs Troolla Tech, posted on X that the agent he uses to screen potential clients needs neither code nor an external framework.

He says he built it directly in Claude Cowork by outlining the steps: search for companies via job postings, LinkedIn and Upwork; review each company's website and LinkedIn page; score each against his defined ideal-client criteria; and log the results in a spreadsheet. He says who he actually reaches out to remains his decision.

In replies on the same post he adds details: one of his rules is to skip any company that already has its own IT people, which the agent checks in the company's LinkedIn staff list before scoring; if the agent cannot check something, the company gets no score; every score in his sheet has to link to its proof, such as the job ad and the company's LinkedIn page; and when sources disagree it records both with their sources, disqualifies the company if either source fails the criteria, and, if it cannot tell which is accurate, gives no score and refers the entry to him for review.

He also says the system logs every gap with the reason, and that when he re-processes the data he conducts a comprehensive review of the entire company rather than patching the specific gaps. He gives no counts, no example sheet and no list of his criteria.

No results are reported: no number of companies screened, time saved or clients won. The only outcome he mentions is one company that came out as a top fit when the agent had never opened its website, which is why he says he added the rule that every score must link to its proof.

Limitations

Short post with no figures, example output, criteria or prompt, so we cannot say how well it works. The author does not say how many companies it screens, how often it runs, how Cowork gets into LinkedIn and Upwork, or how accurate the scores are; the extra rules above come from his own replies.

His main message is that no framework is needed ("Why on earth would you even need a framework?"), and he sells AI and fractional-CTO services, so this is his own account of his own setup. The model is not stated.

Try it in your business

What you'll need

Claude Cowork; your written ideal-client criteria and any disqualifying rules; access to the sites it searches (the post does not say how); a spreadsheet for the results; your own decision on who to contact.

The author says the agent is built from plain written steps inside Claude Cowork with no code and logs to a spreadsheet. The post does not name any connector, and reading LinkedIn and Upwork may need his accounts or browser access, which it does not say. The label is our reading, not something the author states.

Starting prompt

The author didn't publish their prompt (read the full post and the author's replies in the X conversation; the steps are summarized but no prompt is quoted), so this is our adaptation.

How ours differs from the original setup: Same execution agent (Claude Cowork).

The author published no prompt, so this is our starting prompt built from his four steps and from the rules he describes in his replies (no score if a check cannot be made, a proof link for every score, record both sources when they disagree and disqualify the company if either fails the criteria, give no score and refer to him only when it cannot tell which is accurate, skip companies with their own IT staff).

The criteria, the scoring scale, the sources and the spreadsheet name are blanks for you to fill in. It adds an explicit no-contact instruction.

Our suggested starting prompt — not the author's original · untested
Screen potential clients for [my business]. My ideal client is: [criteria]. Skip any company that [disqualifying rule, for example already has its own IT staff]; check that before you score. Find companies through [job postings, LinkedIn, Upwork or other places I name] that look like [type of company or role]. For each company, read its website and LinkedIn page, then score it from [low] to [high] against my criteria. For every score, record a link to the evidence behind it. If you cannot check something, give the company no score and write down what is missing and why. If two sources disagree, record both with their links; if either source fails my criteria, disqualify the company; if you cannot tell which source is accurate, give the company no score and flag it for my review. Put everything in a spreadsheet called [name]. Do not contact any company or send anything; I decide who to reach out to.

Replace the [bracketed] parts with your own files and tools.

Get examples like this every Tuesday and Friday

Practical ways to put AI agents to work in your business. We check the original sources, explain the setup and point out the limits. Free.

Unsubscribe any time. Privacy policy

Report a correction

Related use cases

Categorize bank transactions, number vouchers, match payments to invoices and draft year-end accounts

Have Claude Cowork do the bookkeeping from a bank statement and spreadsheet, match payments to invoices and draft year-end accounts, as the author describes in a LinkedIn post that he says Claude wrote.

Claude Cowork · Finance and bookkeeping · Prompts and files · Source: Andrew Blumson, founder of Adevious, a UK AI training and software business (per adevious.co.uk); the business whose bookkeeping the post describes is not named · October 24, 2026

Draft late invoice reminder emails every Monday for your approval

Have Claude Cowork set up a Monday-night agent that drafts late-invoice reminder emails for you to review and send, as the author describes.

Claude Cowork · Finance and bookkeeping · Connected apps · Source: Michel Abood (San Diego), works in a wine-import business (per his own reply on the post); the posts do not name the company; Rundown profile 'michel' links to linkedin.com/in/michelabood; company size not stated · October 13, 2026

Check local prospects' email security and draft personalized outreach for your approval

Have Grok Bot research local businesses' public email-security records and draft personalized first-touch emails for your approval, the work the author says his AI team did.

Grok Bot · Sales · Custom development · Source: Eddy Kawira, runs a one-person local tech services business in South Bend, Indiana (Eddy Services); infrastructure engineer and writer of the eddykawira.com blog (per the post, his About blurb and his services site) · October 11, 2026