what-to-say-to-this-company
Skill5 starsChanged 4 months agoMITView on GitHub
Turns one target company into one specific opening angle, built from what that company visibly does rather than from a template with the company name pasted in. Use when an account is on the list but nobody knows what to write, or when every outbound email sounds the same.
In this repository30 other files it gives its agents.
- AGENTS.mdAGENTS.md
- ad-to-landing-promise-matchSkill
- b2b-audience-signal-auditSkill
- cold-outbound-sequence-reviewSkill
- competitor-lead-angle-reviewSkill
- crm-lead-source-quality-auditSkill
- discovery-call-gap-analysisSkill
- disqualification-reason-minerSkill
- email-nurture-sequence-reviewSkill
- form-friction-finderSkill
- ghosted-after-the-demoSkill
- google-ads-lead-quality-auditSkill
- icp-reality-checkSkill
- landing-page-qualification-reviewSkill
- lead-magnet-offer-fitSkill
- lead-scoring-sanity-checkSkill
- lead-to-opportunity-drop-off-reviewSkill
- linkedin-ads-lead-quality-auditSkill
- meeting-prep-five-minutesSkill
- meta-ads-lead-quality-auditSkill
- objection-cheat-sheetSkill
- pipeline-hygiene-auditSkill
- proposal-win-loss-reviewSkill
- sales-follow-up-speed-auditSkill
- scale-readiness-check-lead-genSkill
- sdr-handoff-brief-generatorSkill
- spam-folder-checkSkill
- weekly-b2b-lead-gen-readoutSkill
- won-deal-to-case-studySkill
- write-the-follow-upSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 10 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Use it for one company at a time.
- Distinctive: line 32
Look for observable facts with commercial meaning: hiring in a function, a pricing model change, a market they entered, a product line they …
- Distinctive: line 35
Write the opener around that one thing.
- Distinctive: line 43
Never invent a trigger event.
- Distinctive: line 44
If the research turns up nothing usable, say so.
- Distinctive: line 79
Do not fabricate company facts, funding, headcount, customers or events.
- Distinctive: line 80
Do not present a guess as something you found.
- Distinctive: line 81
Do not scrape or store personal data beyond what is needed to write one message.
- Distinctive: line 82
Do not produce bulk personalisation.
- Distinctive: line 83
Do not claim the message will get a reply.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
What to Say to This Company
Use this skill when
The account is chosen and the message is not.
This is the point where most outbound quietly dies. The list is fine, the targeting is fine, and then someone opens a blank email, cannot find anything specific to say, reaches for the template, and sends the same paragraph that the other forty vendors sent that quarter.
Use it for one company at a time. It is not a list-processing tool, and running it across two hundred accounts to generate two hundred openers produces exactly the generic output it exists to avoid.
Required input
- The company name and website.
- Who you sell to there, by role.
- What you actually do, in one honest sentence.
Anything else that is genuinely available helps and none of it is mandatory:
- their careers page, pricing page, or recent announcements
- the specific person's profile or posts
- whether anyone at your company has spoken to them before
- what happened with similar companies you have won or lost
Analysis workflow
- Establish what this company visibly does, in their own words, before forming any opinion about what they need.
- Look for observable facts with commercial meaning: hiring in a function, a pricing model change, a market they entered, a product line they dropped, a job ad that describes the problem you solve.
- Separate the observable from the assumed, and keep them separate in the output. "They posted three SDR roles" is observable. "They are struggling with pipeline" is a guess dressed as insight, and buyers can hear the difference immediately.
- Form one hypothesis about a problem they plausibly have, tied to something specific you found, and state what would disprove it.
- Write the opener around that one thing. One angle, not three.
- Sanity check it with the substitution test: replace the company name with a competitor's name. If the message still works, it is a template and it goes back to step 2.
- Give the honest fallback. Sometimes there is nothing specific to find, and the right move is a short, plain, low-claim message rather than a manufactured observation.
Decision rules
- One angle per message. Two angles read as a pitch, three read as a brochure.
- Observable beats clever. A boring true fact about their business outperforms a smart guess about their strategy.
- Never invent a trigger event. Fabricating a funding round, a hire or a launch is the fastest way to lose an account permanently, and the buyer always knows.
- If the research turns up nothing usable, say so. "No specific angle available, send the short version" is a valid and useful answer.
- Congratulating someone on something public gives you a subject line and nothing to say after it.
Output format
What is actually observable
Bulleted facts with where each came from. Nothing inferred in this section.
The hypothesis
One paragraph: what you think is true about their situation, why, and what would prove you wrong.
The opener
The message, ready to send, short.
The substitution test
State whether the message survives having the company name swapped, and if it does, what was too generic.
If the hypothesis is wrong
The one-line pivot for the reply that says "that is not really our situation."
Practical example
A logistics software company is on the list. The site says nothing useful. But their careers page has two open roles for customer support in a company of forty people, and the job description mentions handling order status questions manually.
The observable fact is two support hires and a manual process named in a job ad. The hypothesis is that support volume is growing faster than headcount can absorb, and that order status is the dominant ticket type. The opener names the job ad, names the specific ticket type, and asks one question about how they handle it today.
The substitution test passes: swap in a competitor and the message breaks, because the competitor is not hiring for that.
Guardrails
- Do not fabricate company facts, funding, headcount, customers or events.
- Do not present a guess as something you found.
- Do not scrape or store personal data beyond what is needed to write one message.
- Do not produce bulk personalisation. This skill makes one message for one company.
- Do not claim the message will get a reply. It removes a reason to ignore you, which is not the same thing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

