← all playbooks

Screen any outreach list against your off-limits rules before it is loaded

A cleared list that is safe to load, a removal report naming the rule and the matched value behind every row taken out, and a short pile of rows the screen could not resolve, left for you to rule on.

use when
a list is about to go into a sequence and you need to be sure no client, no off-limits company, no one in process and no past opt-out is on it
starts from
A list

The prompt

Paste it into Claude Code or the Claude desktop app with Hyreflow connected. The first line tells your agent to use Hyreflow, so it reads the play, asks before it spends anything, and hands the work back to you.

paste this into Claude
Use hyreflow and screen this list against my do-not-contact rules before any
of it goes into a sequence.

The list: <FILE_OR_LIST>. My rules, kept as separate lists:
- Current clients: <CLIENT_LIST_OR_READ_FROM_CRM>
- Companies under an off-limits agreement: <OFF_LIMITS_COMPANIES>
- People already in process with us: <IN_PROCESS_PEOPLE_OR_JOBS>
- Previous opt-outs: <OPT_OUT_FILE>
- Competitors I never write to: <COMPETITORS>

How I want it run:
- Match companies on website domain and known name variants, never on a
  name alone. Match people on LinkedIn profile or email address.
- Delete nothing. Give me the cleared list, plus a removal report with the
  rule and the matched value on every row.
- Anything you cannot resolve goes on a third list for me to rule on.

Screen five rows first and show me. The screen itself should be free. If
rows with no website need a lookup, tell me what the full run costs before
you do the rest. Do not load or send anything.

Replace every <PLACEHOLDER> with your own detail. Everything else can stay as written.

What you need first

  • The list you are about to contact, with a company website, a LinkedIn profile or an email on each row
  • Your do-not-contact rules as lists: clients, off-limits companies, people in process, opt-outs, competitors
  • Optional: a connected recruiting CRM, so client companies and live pipelines can be read from it

Tools it can reach for

The agent picks per step from what your workspace has. Nothing here is required by name.

What happens when you run it

Free steps are marked free. Anything that spends credits is marked, and the agent asks before the first paid run of any size.

  1. 1

    Collect the rules as separate lists

    free

    Clients, companies under an agreement, people in process, opt-outs and competitors are kept apart, because each one blocks for a different reason and the report has to say which rule caught a row. A rule you have not supplied is named as not applied. It is never quietly assumed.

  2. 2

    Read what your CRM already holds

    free

    With a recruiting CRM connected, company records can be read from it, and Recruit CRM can return the companies flagged off limits there. People in process are read from the pipeline of the jobs you name, where the CRM exposes it. Reading your own system costs no credits.

  3. 3

    Read opt-outs where the sender keeps them

    free

    SendKit holds a do-not-contact list and lemlist records unsubscribe events, and both can be read. For any other sequencer you export the unsubscribes and hand over the file.

  4. 4

    Resolve company names to domains

    credits

    Only for entries that arrive as a name with no website. One web search per company finds the domain and the names it trades under. This is the only paid part, and a search is charged whether or not it finds anything.

  5. 5

    Run the screen

    free

    Every row is compared by company domain first, then by known name variant, and people by LinkedIn profile and email address. The domain of an email address counts as evidence of the employer. It is a comparison on data already in hand, so no provider is called.

  6. 6

    Set aside what cannot be resolved

    free

    A row with no domain, a company name that half matches an off-limits entry, or a person matched on name only goes on a separate list with the reason. It is neither cleared nor removed until you rule on it.

  7. 7

    Report every removal

    free

    The cleared list, the removal report with the rule and the matched value on each row, and the counts per rule. Your source file is left untouched, and nothing is loaded into a sequencer.

What a run costs

Credits are spent per record the play actually works, and a lookup that finds nothing usually costs nothing. The two figures are the run where the first provider answers and the run where every lookup walks its full chain.

recordsif the first provider answersif every lookup walks the chain
25under $12.5 creditsunder $12.5 credits
100$110 credits$110 credits
500$550 credits$550 credits
1,000$10100 credits$10100 credits

Free before anything is charged

  • Collect the rules as separate lists
  • Read what your CRM already holds
  • Read opt-outs where the sender keeps them
  • Run the screen
  • Set aside what cannot be resolved
  • Report every removal

What moves the number

  • How many records survive the free filters. Everything dropped before the paid steps costs nothing.
  • Providers you connect with your own key. Those calls bill your account, not your credits.

An estimate, not a quote, priced at the volume credit rate. Your agent sizes the run against your own workspace and tells you what it will cost before it spends anything.

The check that protects the fees you already earn

Most outreach mistakes are survivable. A weak subject line costs a reply. Writing to the head of talent at a current client to ask whether her engineers are open to a move costs the account. So does approaching someone who is at second interview through your own colleague, or emailing a person who asked to be left alone last spring.

Every agency has these rules. They tend to live in a partner's head, a spreadsheet nobody has opened since January, and a flag in the CRM that half the team uses. The check itself is a recruiter scrolling a list by eye the evening before a send. This play makes it a gate with a paper trail: any list, from any source, passes through it before it goes near a sequencer.

It is not a sourcing play. Sourcing from named competitors builds a shortlist with the screen folded in. This one takes a list that already exists, whether it came from a search, an event, an export or a colleague, and clears it for outreach.

What you get back

Three files in your working folder. The cleared list, in the same columns it arrived in. The removal report: every row taken out, the rule that caught it, and the value it matched on, such as the domain or the profile link. And the unresolved pile, with the reason each row could not be settled.

The counts per rule sit at the top, so a list that lost a third of its rows to one rule is obvious at a glance.

Variations worth knowing

Block and flag are different. A current client blocks outright. A prospect another consultant is working may only need a flag. Say which rules remove a row and which annotate it.

Screen in both directions. On a candidate list, the client and agreement rules apply to the current employer. On a business development list, they apply to the company you are pitching.

Carry it forward. The cleared list is a ready input for building an outreach sequence, and the removal report is what you show a client who asks how their people are kept safe.

Where this goes wrong

A rule nobody supplied. The screen cannot enforce an agreement it was never told about. Missing lists are named in the report as not applied.

Personal email addresses. A personal address says nothing about the employer, so candidate rows are matched on the employer field and the profile link. If those are stale, the match is too.

Opt-outs spread across tools. Someone who unsubscribed in a sequencer you left last year is only caught if that export is in the file you hand over.

A job move nobody recorded. A row that still shows the old employer clears the screen. Check the top of a candidate list by eye before the first message goes.

Questions

Why is a name match not enough?

Because companies do not spell themselves one way. A group brand, a regional entity and an abbreviation can all be the same client, and two unrelated firms can share a name. A website domain is far harder to confuse, so the screen matches on domain first and uses name variants as the second check. What it cannot settle is listed for you. It is never passed through.

Does the screen cost anything?

The comparison itself is free, because it works on rows you already hold and lists you already gave. Reading your own CRM or sequencer costs no Hyreflow credits either. The one paid part is finding a website for a company that arrived as a name only, and you see what that comes to before it runs.

Can it read my client list from the CRM instead of me typing it?

Where a recruiting CRM is connected, yes. Company records can be read from Recruit CRM, Loxo, JobAdder, Recruiterflow and Atlas. A CRM holds prospects as well as clients, so you say which status, owner or flag marks a live client, and the agent shows you the resulting list before it is used.

What about parent companies and subsidiaries?

Ownership is not in the data on a contact list. If a client's sister company or parent is also off limits, name it in the rules. The screen is best effort on what it has been told, and it says so in the report.

Does it delete the blocked rows from my list or my CRM?

No. It writes a cleared copy and a removal report into your working folder and leaves the source alone. Hyreflow has no delete for CRM records at all, so nothing is removed from your system either.